Women’s Perceptions and Experiences of Domestic Violence: An Observational Study From Hyderabad, Pakistan

Abstract

This community-based observational study of 1,325 women seen for antenatal care examined how women in Pakistan define violence against women (VAW), with an emphasis on domestic violence, what an acceptable response to violence is, reasons for remaining silent, and whether participants are willing to disclose incidents of domestic violence to others. Nearly half of the study participants believed that physical violence was VAW. Verbal abuse, controlling behavior by the husband, conflict with in-laws, overburdening domestic work, and threatening to leave or remarry were also considered VAW. However, only five respondents (0.4%) considered sexual abuse to be VAW. Most women who screened positive for domestic violence responded by remaining silent or verbal fighting back. None sought professional help. Women who decided to remain silent feared that the abuse would escalate or that responding would not help them. Women cited social stigma and concerns about the impact of the violence on children as reasons for not disclosing violent incidents to others or seeking professional help. Women’s lack of autonomy further reduced their ability to take steps against violence. Although societal norms, particularly patriarchal beliefs and women’s subordination to men, likely explain women’s tolerance of abuse, their recognition of physical abuse as violence indicates that they do not necessarily believe it is always justified. Educational interventions to drive changes in the social norms around gender violence along with effective and enforceable legal measures are likely required to ensure women’s safety.

Introduction

Violence against women (VAW) is a global phenomenon that cuts across all cultures, religions, and socioeconomic groups and touches on fundamental issues of power, gender, and sexuality. The United Nations General Assembly (UNGA; 1993) “Declaration on the Elimination of Violence Against Women” defines VAW as

any act of gender-based violence that results in, or is likely to result in, physical, sexual or psychological harm or suffering to women, including threats of such acts, coercion or arbitrary deprivation of liberty, whether occurring in public or in private life. (p. 3)

Domestic VAW is a form of VAW that encompasses “physical, sexual and psychological violence occurring in the family, including battering, sexual abuse of female children in the household, dowry-related violence, marital rape, female genital mutilation and other traditional practices harmful to women” (UNGA, 1993, p. 3).

VAW, particularly intimate partner violence (IPV), is known to adversely affect victims’ physical and mental health and quality of life. The World Health Organization’s (WHO) Multi-Country Study on Women’s Health and Domestic Violence Against Women found that women who had experienced IPV in their lifetimes had poorer self-reported health, were more likely to recently have had pain and difficulty performing daily activities, and had greater emotional distress and suicidal thoughts than women who had not been abused (Ellsberg, Jansen, Heise, Watts, Garcia-Moreno, & World Health Organization Multi-Country Study on Women’s Health and Domestic Violence Against Women Study Team, 2008). Women experiencing IPV during pregnancy are more likely to have postnatal depression (Ludermir, Lewis, Valongueiro, de Araujo, & Araya, 2010) and are at greater risk for adverse pregnancy outcomes, including preterm birth and low birth weight (Murphy, Schei, Myhr, & DuMont, 2001; Rodrigues, Rocha, & Barros, 2008) and infant and maternal mortality (Chibber & Krishnan, 2011).

VAW is thought to occur at high rates worldwide and potentially modifiable cultural factors may play an important role in determining both actual rates of violence and attitudes toward its acceptability. In Western countries, domestic and marital VAW is addressed through governmental policies and frameworks. In many Islamic countries, however, several forms of domestic violence are typically not considered a major concern despite their frequency (Douki, Nacef, Belhadj, Bouasker, & Ghachem, 2003). A number of studies have reported varying perceptions of domestic violence in different cultures. For instance, American Indian women defined physical abuse only as domestic violence, whereas European American women tended to include verbal and emotional acts along with physical abuse as domestic violence (Tehee & Esqueda, 2008). Chinese adults defined domestic violence as physical or sexual acts of aggression between spouses as opposed to psychological aggression (Yick & Siewert, 1997). Some studies have found that women have been socialized to believe that men are their guardians, and that men have the right to beat them if they behave unacceptably (Jejeebhoy & Sathar, 2001; Schuler, Hashemi, Riley, & Akhter, 1996). In such contexts, social norms around gender have normalized VAW, particularly domestic violence, to the point where it is believed to be necessary, is construed as a sign of love or affection, and victims are blamed for violence that is perpetrated against them by their partners (Amoakohene, 2004; Douki et al., 2003; Garcia-Moreno, Jansen, Ellsberg, Heise, & Watts, 2005; Haj-Yahia, 2002; Kulwicki & Miller, 1999).

Very few women seek formal or legal help to escape violence, particularly domestic and IPV (Chandra, Deepthivarma, Carey, Carey, & Shalinianant, 2003; Douki et al., 2003; Jejeebhoy & Bott, 2003; Panchanadeswaran & Koverola, 2005; Raj & Silverman, 2002; Shaikh, 2000). A report from the WHO found that 55% to 95% of women, when faced with violence, did not seek help from formal organizations or people in positions of authority (Garcia-Moreno et al., 2005). A study of South Asian women living in Boston (USA) found that only 11.3% of the women experiencing violence perpetrated by their intimate partner reported receiving counseling, and only 6.9% visited a doctor for consequences of the abuse (Raj & Silverman, 2002). In Bangladesh, 66% of women remained silent in response to physical abuse by their husband. The main reasons behind the silence were societal normalization of violence, stigma, and fear of greater harm (Naved, Azim, Bhuiya, & Persson, 2006). In Asian cultures, patriarchal norms and the pressure to maintain harmony, peace, and family honor force women who are abused into staying silent because the prospect of facing violence is preferable to the social stigma and potential isolation associated with reporting abuse to authorities external to the family (Panchanadeswaran & Koverola, 2005). Thus, VAW is widely overlooked and is deeply embedded in many cultures.

In Pakistan, where a strong patriarchal belief system dominates societal norms and behaviors, VAW is expected to be highly prevalent (Fikree & Bhatti, 1999; Raj & Silverman, 2002; Shaikh, 2000). A recent systematic review found that the prevalence of physical IPV perpetrated by a husband against his wife ranges from 16% to 80%; psychological violence ranges from 48% to 84% (P. A. Ali, Naylor, Croot, & O’Cathain, 2014). The prevalence of sexual violence against pregnant women ranged from 14% to 21% (P. A. Ali, Naylor, et al., 2014). Moreover, patriarchal ideas contribute to women’s social, economic, and emotional dependence on their husbands. This may prevent women from leaving their abusive husbands. Furthermore, cultural practices and social norms may lead to the belief that VAW is acceptable and thus prevent women from disclosing their abuse experiences (Andersson et al., 2010). Responses to victimization by women who are abused are influenced by their views of themselves, societal norms, and the attribution that they make for the causes of abuse.

The aim of this study was to explore how women in Pakistan generally define VAW and how they respond to violence when they face perpetration. The study also focused on women’s willingness to disclose incidents of violence to others as well as the reasons why they would not disclose VAW. Research on attitudes in Pakistan toward IPV in particular and VAW in general has yet to be conducted (P. A. Ali, Naylor, et al., 2014). This article aims to explore these issues and add to the literature on cultural norms regarding gender and violence in Pakistan, providing context for service providers to implement appropriate interventions aimed at reducing the acceptability of VAW in this patriarchal society. Because of cultural norms and societal expectations of women, we hypothesized that Pakistani women do not define violence as it is defined by UNGA (1993), which may contribute to greater acceptance of abusive behaviors. We also hypothesized that when faced with violence, most women would choose to remain silent because of cultural norms and the belief that no help could be sought.

Method

We conducted a cross-sectional study with few open-ended questions of pregnant women in Hyderabad, an Urdu-speaking city in Pakistan of about 1 million inhabitants. This study was part of a larger study that examined a range of sociodemographic, psychosocial, nutritional, and clinical factors associated with infectious morbidity and pregnancy outcomes among mothers and infants. Details related to methodology and further results are described elsewhere (Karmaliani et al., 2008). The study was approved by the Ethics Review Committee at the Aga Khan University and the Institutional Review Boards of the University of Alabama, Birmingham, and Research Triangle Institute International. The WHO ethical and safety guidelines for research in domestic violence were followed. Lady Health Workers (LHWs) were hired, trained, and certified to conduct interviews and to ensure that participants attended the study clinic for periodic clinical evaluations. LHWs in five selected units of the Hyderabad community screened 1,879 participants during their routine prenatal home visits. Of these participants, 1,774 (94%) gave verbal consent to attend the study to confirm eligibility. Based on a clinic visit, 1,376 of these women met the inclusion criteria, and 1,368 gave informed consent to participate in the parent study while at the clinic. Of these, 1,325 (96%) completed the domestic violence questionnaire.

Participants

Women in their 20th to 26th weeks of pregnancy (confirmed by ultrasound), who were permanent residents (i.e., intending to stay in Hyderabad for a minimum of 6 months) and able to give verbal and written consent, were included as study participants. Only one participant from each household was enrolled even when other women met study inclusion criteria. This minimized the risk of disclosure and harm to women’s safety because other family members would not know what was discussed with the participant (WHO, 1999). Participants who did not meet the inclusion criteria had a clinical diagnosis of a life-threatening condition and/or planned to deliver outside of the project area were excluded from the study.

Study Procedures and Measures

Participants were interviewed in the research clinic without any family members present. Participants were screened for domestic violence and asked a series of open- and closed-ended questions about their experiences regarding domestic violence using a WHO screening instrument modified based on the Pakistani National Gender Indicators List for Violence Against Women. Qualitative questions regarding women’s perceptions of VAW and their experiences of remaining silent in response to violence are the primary focus of this article. The questions were primarily broad and open ended to ensure that women were free to express their views so that they were not influenced by prompts, and we could obtain an accurate representation of women’s beliefs about VAW. All women were asked what they believed encompassed VAW, although only women who indicated that they had experienced or were experiencing violence were asked about how they responded to the violence (closed ended) and why (open ended). A demographic questionnaire, which collected data about the participant’s age, employment status, and educational status, as well as the husband’s employment status, was also included in the study.

The primary outcome of this study was to identify what broad categories of violence (e.g., psychological, economic, physical, sexual) were considered by participants to comprise VAW. This is reported as the proportion of women that believed a certain type of violence to comprise VAW. Another outcome was to identify what proportion of women took action when faced with violence. Among those who remained silent, we aimed to identify what the most common reasons for doing so were.

Participants were informed that they could withdraw from the study at any point, that they could choose not to answer any or all of the questions, that the interviewer would take notes, and that their responses would be kept confidential. Actions to prevent potential distress and referral for support through counseling were ensured.

Data Analysis

The qualitative responses were reviewed and common themes identified using content analysis (Krippendorf, 2012) to develop a coding framework. An iterative process was then used for coding responses based on this framework to ensure interrater reliability. First, one author (C.P.) coded the responses into categories based on the themes. The majority of responses were brief and unequivocally fell under one of the themes identified (e.g., “beating,” “yelling,” or “does not give me money for food”). A second author (R.K.) then reviewed the coding. Any discrepancies or uncertainties in coding were discussed among the two coders to reconcile differences and achieve consensus. Finally, to ensure study rigor, the primary author (F.I.M.) then reviewed the content analysis and recoded the qualitative responses; based on this review, in agreement with the other coders, minor revisions were made to the coding framework to improve clarity.

Responses to the question regarding what encompasses VAW were grouped into seven categories: (a) physical abuse, (b) verbal abuse, (c) sexual abuse, (d) controlling behaviors, (e) conflict with in-laws, (f) bad behavior, and (g) emotional abuse. The category “bad behavior” encompassed vague responses that were not probed further, and simply noted by the LHWs as “husband’s bad behavior” or simply “bad behavior.” It is presumed that the “bad behavior” referred to by participants would encompass behaviors or actions taken by an intimate partner, family member, or other person that would cause a woman (presumably the respondent) psychological or physical harm; however, without further probing, this cannot be confirmed. We included this category because the content analysis found a large proportion of respondents mentioned “bad behavior.”

A large number of participants choose to “stay quiet” in response to violence. Reasons for this were similarly grouped into 10 categories: (a) fear of escalating violence, (b) respect for elders/in-laws, (c) no one to confide in, (d) husband will leave/make her leave, (e) won’t make a difference/helplessness, (f) for the sake of children, (g) hope abuse will stop, (h) husband’s right, (i) self-respect, and (j) own fault. An additional “other” category highlighted responses that could not be categorized otherwise.

Frequencies were calculated for all the categories. In addition, chi-square tests and odds ratios (ORs) were computed to compare perceptions of and responses to VAW across demographic groups. Responses of open-ended questions were coded into categories using Microsoft Excel and were imported into SAS (v. 9.3) for quantitative analysis.

Results

The demographic characteristics of the participants and their perceptions of domestic violence are shown in Table 1. The average age of the participants was 27 years, and two thirds (n = 886) had received some formal education. Seventy-eight percent (n = 1,038) of the participants’ husbands had permanent employment, and on average, families owned about half of the five items included in the property index. Four percent (n = 55) of the women reported that their husbands had other wives or sexual partners. Twenty-eight percent (n = 375) of women were younger than 18 years old when they first had sexual intercourse, and 82% (n = 1,087) had previous pregnancies. Six hundred eighty-one women (51%) reported experiencing some form of violence before and/or during pregnancy.

Table 1.

Sociodemographic Characteristics of Participants (N = 1,325).

Characteristic	n	%
Age
20 or less	162	12
21–25	458	35
26–30	449	34
More than 30	256	19
Education
None	439	33
<6 years	244	18
6+ years	642	48
Employed
Yes	167	13
No	1,156	87
Husband has a permanent job
Yes	1,038	78
No	285	22
Property index
Low	592	45
High	729	55
Experienced domestic violence before and/or during pregnancy
Yes	681	51
No	644	49
Husband has other wives/sex partners
Yes	55	4
No	1,270	96
Age at first intercourse
<18 years	375	28
18+ years	950	72
Ever pregnant before
Yes	1,087	82
No	237	18
Wanted this pregnancy
Yes	819	62
No	505	38

Note. Missing data by characteristic: employment (n = 2), husband’s job (n = 2), property index (n = 4), ever pregnant (n = 1), and wanted pregnancy (n = 1).

The majority of the women interviewed referred to some sort of domestic violence as VAW (Table 2), identifying different types of domestic violence as VAW. Forty-seven percent (n = 616) of women stated that they thought physical abuse by a husband, including beating and burning, was VAW, whereas 8% (n = 109) thought that verbal abuse, including being critical, using abusive language, teasing, taunting, and scolding, was VAW. Only five respondents (0.4%) mentioned sexual abuse as VAW. Controlling behaviors by the husband, such as not giving the wife money or food, preventing her from going out of the house to work or visit relatives, and overburdening her with domestic work, were considered VAW by 21% (n = 279) of the women. About 22% (n = 289) of women also explicitly stated that any kind of conflict with in-laws (physical, verbal, or emotional, but not sexual) was VAW. Emotional abuse by a husband or in-laws, comprising behaviors such as disrespecting women, not listening to her or giving her any importance, threatening to leave or remarry, and always being suspicious of her, were considered to be emotional abuse by 28% of respondents. In addition, a number of women stated that abuse stemming from the inability to have children would constitute VAW. Some women were not specific in their replies; 42% (n = 557) described VAW as “bad behavior,” “domestic violence,” “fighting,” “quarreling,” “being aggressive,” or “having a bad attitude.” These responses usually referred to the husband’s actions and were grouped under one category: bad behavior/domestic violence.

Table 2.

Percentage of Women Indicating a Type of Abuse Is Violence Against Women, by Demographic Characteristics.

	Type of Abuse Believed to Be Violence Against Women
	Physical	Verbal	Sexual	Controlling Behavior	Conflict With In-Laws	Bad Behavior/DV	Emotional
Characteristic	n (%)	n (%)	n (%)	n (%)	n (%)	n (%)	n (%)
Overall	616 (47)	109 (8)	5 (0)	279 (21)	289 (22)	557 (42)	366 (28)
Age
20 or less	81 (50)	12 (7)	0 (0)	32 (20)	47 (29)	61 (38)	32 (20)*
21–25	201 (44)	40 (9)	1 (0)	101 (22)	106 (23)	181 (40)	144 (31)
26–30	201 (45)	36 (8)	2 (0)	97 (22)	96 (21)	199 (44)	118 (26)
More than 30	133 (52)	21 (8)	2 (1)	49 (19)	40 (16)	116 (45)	72 (28)
Education
None	225 (51)*	32 (7)	3 (1)	89 (20)	89 (20)	213 (49)***	90 (21)***
<6 years	118 (48)	20 (8)	0 (0)	40 (16)	59 (24)	107 (44)	66 (27)
6+ years	273 (43)	57 (9)	2 (0)	150 (23)	141 (22)	237 (37)	210 (33)
Employed
Yes	74 (44)	15 (9)	0 (0)	38 (23)	36 (22)	80 (48)	40 (24)
No	541 (47)	93 (8)	5 (0)	240 (21)	253 (22)	477 (41)	326 (28)
Husband has a permanent job
Yes	473 (46)	91 (9)	5 (0)	224 (22)	220 (21)	424 (41)	308 (30)**
No	142 (50)	18 (6)	0 (0)	55 (19)	69 (24)	132 (46)	58 (20)
Property index
Low	307 (52)***	44 (7)	1 (0)	117 (20)	131 (22)	245 (41)	141 (24)**
High	308 (42)	65 (9)	4 (1)	160 (22)	158 (22)	309 (42)	224 (31)
Experienced domestic violence before and/or during pregnancy
Yes	313 (46)	67 (10)*	4 (1)	160 (23)*	166 (24)*	272 (40)	198 (29)
No	303 (47)	42 (7)	1 (0)	119 (18)	123 (19)	285 (44)	168 (26)
Husband has other wives/sex partners
Yes	23 (42)	6 (11)	0 (0)	14 (25)	8 (15)	27 (49)	21 (38)
No	593 (47)	103 (8)	5 (0)	265 (21)	281 (22)	530 (42)	345 (27)
Age at first intercourse
<18 years	191 (51)	26 (7)	1 (0)	83 (22)	83 (22)	176 (47)*	100 (27)
18+ years	425 (45)	83 (9)	4 (0)	196 (21)	206 (22)	381 (40)	266 (28)
Ever pregnant before
Yes	508 (47)	89 (8)	3 (0)	223 (21)	229 (21)	472 (43)*	300 (28)
No	107 (45)	20 (8)	2 (1)	56 (24)	60 (25)	85 (36)	66 (28)
Wanted this pregnancy
Yes	366 (45)	69 (8)	3 (0)	167 (20)	187 (23)	347 (42)	222 (27)
No	250 (50)	40 (8)	2 (0)	112 (22)	102 (20)	209 (41)	144 (29)

Note. These analyses include the 1,314 women who responded to the question about definition of violence against women.

DV = domestic violence.

^*p < .05.

^**p < .01.

^***p < .001.

Table 3 shows that the majority of women (55%, n = 360) reported verbally fighting back in response to verbal abuse, whereas 64% (n = 417) remained silent. In response to physical abuse, 42% (n = 68) of women verbally fought back, and 48% (n = 78) remained silent. The majority (73%, n = 117) of women remained quiet in response to sexual abuse. Twenty-eight respondents (4%) returned to their parents’ home in response to verbal abuse, and 16 respondents (10%) returned after physical abuse. Very few respondents spoke with friends or family, and none consulted professional organizations or took legal action.

Table 3.

How Women Responded to Domestic Violence.

	Verbal Abuse (n = 655)	Physical Abuse (n = 163)	Sexual Abuse (n = 160)
Response	n (%)	n (%)	n (%)
Physically fought back	NA	9 (6)	3 (2)
Verbally fought back	360 (55)	68 (42)	66 (41)
Kept quiet	417 (64)	78 (48)	117 (73)
Returned to parents	28 (4)	16 (10)	0 (0)
Attempted suicide	6 (1)	6 (4)	1 (1)
Talked to family/friends	2 (0)	1 (1)	0 (0)
Took legal action	0 (0)	0 (0)	0 (0)
Consulted with a professional organization	0 (0)	0 (0)	0 (0)
Other	17 (3)	16 (10)	7 (4)

Women who indicated that they responded to abuse by remaining quiet were asked to describe their reasons for doing so (Table 4). The women in this study cited a wide range of reasons for staying in violent relationships, the most common of which was fear that the violence would escalate with almost half (49.4%) of women stating this. Several also cited respect for their elders. Respondents cited a lack of hope that responding to violence would change their situation, stating, “there is no point in talking to him,” “there is no point talking about it every day as it will have no effect on his behavior,” “no use of asking for rights,” “no use in replying,” and “it makes no difference.” A few respondents (15%) indicated that they remained silent in response to verbal abuse out of respect for their elders and in-laws, stating: “I don’t want to be disrespectful towards my elders,” “I am weak and cannot do anything,” and “I am powerless.” Women also revealed that they had no one to confide in, especially those who remained quiet about physical and sexual abuse. A small percentage (12%, n = 39) of women who remained silent in response to all three types of abuse did so for the sake of the children and/or feared that their husbands would leave or make her leave. Self-blame was also evident, with 3 participants stating that the “abuse is due to our own fault.” Of the 73% of women who remained quiet in response to sexual abuse, 22% (n = 16) stated that it was their husband’s rights to have sex with them. Fourteen respondents (19%) felt helpless when faced with sexual abuse, and a significant number of sexual abuse victims (40%, n = 29) feared that violence would escalate. Six women remained silent in response to the three different types of abuse to maintain self-respect. One woman stated that she chose to remain silent because “there is anger in silence.”

Table 4.

Reasons Why Women Stayed Quiet in Response to Domestic Violence.

	Verbal Abuse (n = 229)	Physical Abuse (n = 38)	Sexual Abuse (n = 73)
Reason	n (%)	n (%)	n (%)
Fear of escalating violence	120 (52)	19 (50)	29 (40)
Respect for elders/in-laws	35 (15)	5 (13)	3 (4)
No one to confide in	11 (5)	4 (11)	2 (3)
Husband will leave/make her leave	21 (9)	4 (11)	6 (8)
Won’t make a difference/helplessness	24 (10)	2 (5)	14 (19)
For the sake of children	6 (3)	2 (5)	0 (0)
Hope abuse will stop	8 (3)	0 (0)	0 (0)
Husband’s right	0 (0)	0 (0)	16 (22)
Self-respect	4 (2)	1 (3)	1 (1)
Own fault	2 (1)	1 (3)	0 (0)
Others	5 (2)	1 (3)	2 (3)

An OR analysis revealed the characteristics of women who were more likely to respond to domestic abuse (Table 5). Women older than 30 years of age (see Table 5) who had had a prior pregnancy (OR = 4.04, 95% confidence interval [CI] = [2.36, 6.90]) were more likely to fight back while women who wanted the pregnancy (OR = 0.65, 95% CI = [0.48, 0.89]) were less likely to fight back in response to verbal abuse. Women who had their first instance of sexual intercourse before the age of 18 (OR = 1.96, 95% CI = [1.00, 3.84]) and/or had a prior pregnancy (OR = 4.24, 95% CI = [1.18, 15.20]) were more likely to respond verbally or physically to sexual abuse. No other significant associations were found.

Table 5.

Percentages of Women Who Fought Back Verbally or Physically in Response to Abuse by Demographic Characteristics.

	Verbal Abuse (n = 655)		Physical Abuse (n = 163)		Sexual Abuse (n = 160)
Characteristic	Fought Back n (%)	OR (95% CI)	Fought Back n (%)	OR (95% CI)	Fought Back n (%)	OR (95% CI)
Age
20 or less	41 (54)	0.47 [0.26, 0.87]*	12 (43)	0.83 [0.26, 2.69]	6 (33)	0.82 [0.25, 2.68]
21–25	108 (48)	0.37 [0.23, 0.59]*	25 (42)	0.76 [0.27, 2.15]	19 (37)	0.90 [0.37, 2.16]
26–30	125 (54)	0.47 [0.29, 0.75]*	28 (50)	1.11 [0.39, 3.15]	27 (51)	1.71 [0.73, 4.01]
More than 30	86 (72)	REF	9 (47)	REF	14 (38)	REF
Education
None	137 (56)	REF	38 (43)	REF	24 (43)	REF
<6 years	66 (52)	0.86 [0.56, 1.33]	17 (55)	1.50 [0.65, 3.46]	17 (43)	0.93 [0.40, 2.12
6+ years	157 (55)	0.98 [0.69, 1.38]	19 (43)	1.00 [0.48, 2.08]	25 (39)	0.85 [0.41, 1.77]
Employed
Yes	45 (49)	0.76 [0.49, 1.18]	10 (43)	0.93 [0.38, 2.26]	8 (33)	0.68 [0.27, 1.71]
No	315 (56)	REF	64 (46)	REF	58 (43)	REF
Husband has a permanent job
Yes	263 (53)	0.78 [0.55, 1.12]	53 (49)	1.64 [0.84, 3.20]	47 (42)	1.11 [0.55, 2.23]
No	96 (59)	REF	20 (37)	REF	18 (39)	REF
Property index
Low	184 (58)	1.26 [0.92, 1.72]	55 (49)	1.72 [0.86, 3.41]	30 (43)	1.16 [0.61, 2.19]
High	174 (52)	REF	19 (37)	REF	36 (40)	REF
Husband has other wives/sex partners
Yes	17 (57)	1.08 [0.52, 2.27]	8 (57)	1.70 [0.56, 5.15]	7 (54)	1.77 [0.57, 5.53]
No	343 (55)	REF	66 (44)	REF	59 (40)	REF
Age at first intercourse
<18 years	126 (60)	1.33 [0.95, 1.86]	36 (52)	1.65 [0.88, 3.10]	27 (52)	1.96 [1.00, 3.84]*
18+ years	234 (53)	REF	38 (40)	REF	39 (36)	REF
Ever pregnant before
Yes	339 (59)	4.04 [2.36, 6.90]*	67 (47)	1.91 [0.69, 5.30]	62 (44)	4.24 [1.18, 15.20]*
No	21 (27)	REF	7 (35)	REF	4 (20)	REF
Wanted this pregnancy
Yes	183 (50)	0.65 [0.48, 0.89]*	33 (39)	0.56 [0.30, 1.04]	35 (43)	1.10 [0.58, 2.06]
No	176 (61)	REF	41 (53)	REF	31 (40)	REF

Note. Percentages are row percentages indicating the percentage of women in the subgroup who fought back verbally and/or physically (e.g., 54% of women aged 20 years or less fought back in response to verbal abuse).

OR = odds ratio; CI = confidence interval; REF = reference category.

^*p < .05.

An analysis of women who stayed quiet in response to abuse revealed similar results (not shown in table), with women aged 21 to 25 years (OR = 2.00, 95% CI = [1.26, 3.16], p < .01) and those aged 26 to 30 (OR = 1.67, 95% CI = [1.07, 2.63], p = .026) being more likely to remain silent in response to verbal abuse compared with women 30 years. Women with basic schooling of Grade 6 and more to be more likely to remain quiet in response to verbal abuse, although the association did not reach significance (OR = 1.43, 95% CI = [1.00, 2.04], p = .052). Women with a low family property index, a measure of socioeconomic status, were more likely to remain quiet in response to physical abuse compared with women with a high family property index (OR = 0.50, 95% CI = [0.25, 0.99]).

Discussion

VAW is commonly perpetrated by a woman’s partner, often in her home, and is typically considered to be a private matter outside of the realm of public debate or exploration. Until recent decades, VAW remained largely hidden and undocumented (WHO, 2000). In our study, most of the women (47%, n = 616) stated that physical abuse is VAW, but 109 (8%) considered verbal abuse to be violence, and only 5 (0.4%) women stated that they believed that sexual abuse is VAW. A likely explanation is that abuse, particularly sexual abuse, is widely accepted among women in South Asian societies, and so there is a high tolerance and acceptance of abuse as being part of life (Chandra et al., 2003). Previous studies have demonstrated that societies dominated by patriarchal ideologies, such as those in Pakistan, are more likely to tolerate VAW, and thus make women more accepting of abuse (Ahmad, Riaz, Barata, & Stewart, 2004; T. S. Ali, Krantz, & Mogren, 2012; Faramarzi, Esmailzadeh, & Mosavi, 2005; Haj-Yahia, 2002; Jewkes, 2002). In fact, it was perceived that men are required to use violence as a means to control their wives’ behavior, a measure that Pakistani women are conditioned to accept as normal (T. S. Ali et al., 2012; Rabbani, Qureshi, & Rizvi, 2008; Zakar, Zakar, & Kraemer, 2011, 2013). A study from Karachi found that two thirds of women interviewed believed a wife should always follow her husband’s instructions irrespective of her will, and 18.5% believed violence was justified if she did not follow her husband’s instructions (P. A. Ali, Naylor, et al., 2014).

An important finding in our study was that only five women (0.4%) stated that they believed that sexual abuse is VAW, and the majority (60%) of those who experienced sexual abuse remained silent. Societal norms dictate that a man controls a woman’s sexuality, and that a married woman is required to be sexually available to her husband at all times (Abraham, 1999). Married adolescents perceive that acquiescing to forced sex by husbands is part of their marital duty and is a show of commitment (Jejeebhoy & Bott, 2003; M. E. Khan, Townsend, Sinha, & Lakhanpal, 1997). In this context, the use of force when a request for intercourse is met with refusal is considered normative behavior (Population Council, 2004). A study in India confirmed the acceptance of sexual abuse as part of marital life; neither men nor women perceived this kind of violence to be wrong (Martin, Tsui, Maitra, & Marinshaw, 1999). Sexual violence has been shown to have an adverse impact on women’s mental and physical health independent of their experience of other forms of violence (Dutton et al., 2006). A study in Pakistan found that almost 76% of women who experienced sexual violence had suicidal thoughts (T. S. Ali, Mogren, & Krantz, 2013). Thus, societal norms that keep sexual violence hidden are likely to contribute to the burden of disease attributable to injuries, sexually communicable diseases, and mental health.

Studies have shown that many women find nonphysical violence to be more devastating than physical abuse (Panchanadeswaran & Koverola, 2005; WHO, 2005). Several studies have documented the detrimental effects of psychological violence, both in Pakistan and globally, due to the internalization of feelings of self-doubt and fear (P. A. Ali, Naylor, et al., 2014; T. S. Ali et al., 2013; Dutton et al., 2006; Lagdon, Armour, & Stringer, 2014; Zakar, Zakar, Mikolajczyk, & Kramer, 2012). Women who experience controlling behaviors are also more likely to experience higher levels of violence (Johnson, 1996). Economic violence (i.e., limiting control over and access to resources) in particular is known to worsen social inequalities and lead to physical and sexual violence (Fawole, 2008). In our study, a significant percentage of women (49%) considered controlling behavior, usually by the husband, to be VAW. Commonly mentioned forms of controlling behavior, such as denial of food and money to pay for household expenses, or the husband threatening to leave them or throw them out of the house, reflect the helplessness of the women and their dependence on their husbands. Male-dominated hierarchies influence the decision-making power in all domestic spheres: economic, social, and sexual. Women are often viewed as a burden to the household, and are not granted the same rights and opportunities as men, particularly regarding access to economic resources (Go et al., 2003; Niaz, 2003; Schuler et al., 1996). Our study found that physically abused women who remain silent in response to the violence are more likely to be of a lower socioeconomic status, highlighting their dependence on their husbands for financial resources.

A sizable proportion of respondents (22%) stated that some kind of conflict with in-laws (whether verbal or physical) was VAW, indicating the role that in-laws play in provoking and perpetrating VAW. In Pakistani society, a woman typically moves into her in-laws’ house to reside with her husband’s family, where women must respect both her husband and her in-laws (Kapadia, Saleem, & Karim, 2009). The movement into the husband’s family increases the chances of conflict with her in-laws over domestic matters, such as performing household tasks, a finding confirmed in a Pakistani study (A. Khan & Hussain, 2008). In fact, women often perceive the presence of inlaws in the household as a contributing factor to the incidence of domestic violence (Fikree & Bhatti, 1999). Two women from our study stated that they believed “teasing from in-laws because of dowry” was abuse, indicating the presence of conflict due to dowry-related matters. In South India, a study found that violence is less likely when women bring bigger dowries at the time of marriage (Srinivasan & Bedi, 2007). A Pakistani study also linked conflict with in-laws to greater risk of experiencing sexual violence, likely because of a lack of social support (Kapadia et al., 2009).

The ways women responded to domestic violence provide further insight into women’s perceptions of violence. Consistent with other studies, most women remained silent (Andersson et al., 2010; Rabbani et al., 2008). Domestic abuse, particularly sexual abuse, is typically considered normal and, therefore, must be received silently (Rabbani et al., 2008; Zakar et al., 2013). The majority of women who remained quiet in response to violence did so because they feared that the violence would escalate. Many appeared to feel helpless, stating that responding actively to violence would not help their situation but rather would put them at risk of enduring greater violence. These responses do not imply that women believe that violence is necessarily justified, but that women are expected to tolerate the violence. This is an interesting and important finding, as it suggests that women accept violence perpetrated against them due to societal expectations of women rather than a belief that VAW is justified. Several studies have noted that even in Pakistan, women who report experiencing any kind of IPV (psychological, physical, and/or sexual) demonstrate higher rates of adverse health outcomes. They are found to have higher rates of poor mental health outcomes such as depression, anxiety, and suicidal ideation, as well as injuries and poor reproductive outcomes such as unplanned pregnancies, poor prenatal care, and pain in the abdomen and/or vagina (N. S. Ali, Ali, Khuwaja, & Nanji, 2014; P. A. Ali, Naylor, et al., 2014; Zakar et al., 2012). This is comparable with the adverse health outcomes experienced by women in other parts of the world, including Western countries (Bonomi, Anderson, Rivara, & Thompson, 2007; Chibber & Krishnan, 2011; Lagdon et al., 2014). This could indicate that even though women may tolerate IPV and VAW is more generally acceptable in Pakistan, the physical and mental health effects among women who experience perpe- tration particularly by a spouse are as detrimental as when they are experienced by women in a less patriarchal culture.

Even though VAW is widely accepted, several women in our study revealed that they responded to abuse verbally or physically. Older women were more likely to respond this way. Of particular note, multigravida women had 4 times the odds of responding to verbal abuse compared with primigravida women, although this may be an indication of the younger age of primigravida women. Older women may feel more established in the household and may feel they are able to assert some authority. Younger women, however, may have lived in their husband’s household for a shorter time and so feel less confident in responding to violence. Interestingly, the association between age and responding to violence verbally or physically did not hold for physically or sexually abused women, which may be the due to the greater severity of violence and greater fear of escalation. Previous studies have documented that IPV experience increases the risk of unwanted pregnancy (Chibber & Krishnan, 2011; Zakar et al., 2012); however, to our knowledge, ours is the first to document an association between responding to unwanted pregnancy and responding to violence in this context.

None of the women from our study consulted with professional organizations or took legal action for any of the abuse, which is similar to findings in other studies (Andersson et al., 2010; A. Khan & Hussain, 2008). Social stigma and minimal social support leave few choices for victims of violence and being separated from their partners would likely bring shame and dishonor to the families (P. A. Ali, Naylor, et al., 2014; Andersson et al., 2010; A. Khan & Hussain, 2008). The literature cites the many reasons why victims of abuse do not disclose abuse to their health care professionals; most commonly reported reasons are fears of retaliation by their partner, shame, humiliation, denial about the seriousness of the abuse, and concern about confidentiality (Gerbert, Caspers, Bronstone, Moe, & Abercrombie, 1999). Formal institutions also tend to be perceived as ineffective in assisting women facing violence (P. A. Ali, Naylor, et al., 2014).

In our study, very few participants talked to family and friends in response to verbal and physical abuse. One respondent stated that she did “not want to let neighbors and others know that there is domestic violence because of my social status.” The literature, however, highlights that irrespective of who perpetrates the violence, women are more likely to speak to their friends, neighbors, or immediate family members than anyone else (Andersson et al., 2010; Goodkind, Gillum, Bybee, & Sullivan, 2003). The lack of consultation with family or friends among our respondents is possibly because women who consult with friends or family were often advised to return home to the abuser and to sort out the matter within the household, or to simply accept the violence passively (Douki et al., 2003; Haj-Yahia, 2002; Jejeebhoy & Bott, 2003; Panchanadeswaran & Koverola, 2005). This lack of social support explains why many women felt that they had no one to confide in about their experiences.

Limitations

Although the open-ended nature of the question allowed for a variety of answers and a way for the study participants to freely respond, many responses were vague (i.e., the category “bad behavior/domestic violence”). Unfortunately, the LHWs were not trained to prompt the participants to clarify women’s views and some data were lost. Even some of the more specific answers (e.g., “beating”) are very broad in meaning. For instance, “beating” could refer to a milder form of physical abuse, such as slapping, or a more severe form, such as kicking. Some women may have perceived certain types of violence to be VAW but may not have mentioned them because of social norms that dictate that they must remain silent about these issues. For example, sexual issues are considered to be very personal and are not to be discussed outside the home. Explicitly asking about specific types of violence may have prompted women to state that they considered it to be violence. Similarly, in terms of responses to violence, there could be a possibility of recall biases. Because pregnancy is supposed to be a joyous period for women, women’s feelings and actual responses may have been replaced with “keeping silent.” Furthermore, pregnant women may even be more fearful about potential consequences of disclosing acts of violence, so accurate responses may not have been obtained.

In-depth qualitative research about the perceptions and responses to violence among pregnant and nonpregnant Pakistani women is necessary to determine what and how women view domestic violence and whether pregnancy influences their perceptions of and responses to violence. Furthermore, participants should be recruited from urban as well as rural areas of Pakistan and among people from different socioeconomic backgrounds, as results may vary greatly between these groups of people. A comprehensive understanding of women’s perceptions and responses to violence and whether the acceptance of violence is linked to sociodemographic factors such as age, income, education level, employment status, number of pregnancies and living children, and childhood experience of abuse will contribute to developing future interventions. Furthermore, the term husbands’ rights could have been explored more in terms of understanding how women in Pakistan define the term. Due to the nature of the parent study, follow-up on the consequences of women’s participation in the study and the effectiveness of referral could not be achieved. Future research should also consider examining the effectiveness of measuring how abused women cope with VAW because this could lead to identifying ways to improve the mental health of these women.

Conclusions and Recommendations

Understanding women’s perceptions of VAW and their responses to being silent will assist in developing interventions to change traditional ideas that make violence acceptable and to assist women in recognizing abusive behaviors and seeking help. Staying silent in response to violence may cause women psychological harm, leading to depression, anxiety, fear, guilt, shame, or self-harm tactics. Tolerance of abuse can result in greater risk of injury and psychological disturbance, as greater acceptance of violence is linked with greater incidence of violence.

Our data suggest that women appear to identify several forms of abuse correctly but appear to tolerate it rather than take action against it due to societal norms. To break the cycle of violence, a change in attitudes and perceptions is needed. Behavioral interventions to change societal norms and create opportunities to decrease women’s economic dependence on men are key. Education can play a central role in shifting the power dynamic between men and women because it provides girls with the skills and ability to earn a livelihood, raising their self-esteem so that they respond to violence with greater confidence. Girls who remain in school longer are less likely to be the victim of abuse (Karmaliani et al., 2008). Creating opportunities for girls to access basic schooling and ensuring gender-equitable access to primary level education at a minimum could empower future generations of women. According to the World Economic Forum (2013), Pakistan, at 135th place among all countries, is one of the worst performing countries in the world in terms of gender inequality. Therefore, knowledge about gender equity, women’s rights, family violence, and positive coping strategies should be incorporated in the basic education at primary and secondary school curricula for both girls and boys.

More broadly, programs that educate women and men about violence should be implemented in an effort to change cultural norms around VAW. These programs should aim to increase awareness about the types of violence, particularly in a domestic context, and the risk they may involve, such as psychological and physical injuries. Sexual violence should be a major component of these programs, as it appears to be the most widely accepted form of abuse. The media can play an important role in educating masses about women’s rights. Enrolling religious leaders to play a role positively in preaching to the people about Islamic values and women’s rights that strongly condemn VAW could be an effective manner of bringing about behavior change.

Women should also be encouraged to seek help. Disclosing information about abuse experiences empowers women to take action against abuse as it ends the silence (McFarlane, 2007). For this to be possible, the establishment of both formal and informal support systems is necessary. International health organizations recommend that health care practitioners follow a protocol to identify and manage victims of abuse (Gazmararian, Peterson, Spitz, Saltzman, & Marks, 2000). Primary health care practitioners in Pakistan rarely have resources to support victims of domestic violence, screening tools to detect victimization, or knowledge of referral services for victims (Zakar et al., 2011). It is therefore important to include violence specific knowledge and skills in the health care practitioners’ curricula. Setting up support groups of local women as a resource for abused women to seek help and advice has also been suggested (Andersson et al., 2010).

Recently, the Parliament of Pakistan has passed a number of laws to enhance women’s rights. However, the extent of their enforcement is limited, and it is believed that the legislation has had little effect on preventing VAW (Gill & Stewart, 2011; Immigration and Refugee Board of Canada, 2013). There is a need for correct understanding and implementation of these laws among communities and law enforcement agencies. Authorities, especially the police, should be trained to help women, as studies have shown that women who sought help from police were more likely to receive necessary health care, and were less likely to be repeatedly assaulted (McFarlane et al., 2005). Unless resources and mechanisms are put in place to enforce the law by protecting victims and taking appropriate punitive action against perpetrators, widespread change in how victims respond to violence is unlikely to occur.

Biographies

Farhana I. Madhani is currently a PhD student at McMaster University School of Nursing, Canada. She is an assistant professor at the Aga Khan University, School of Nursing and Midwifery (AKU SONAM), Karachi, Pakistan. She graduated with master’s in science from University of Wisconsin–Madison and baccalaureate from AKU SONAM. She is interested in women’s mental health, gender-based violence, methodological issues in conducting violence against women studies, poverty and mental health, and nursing education specific to clinical teaching. Her PhD thesis focuses on women’s mental health in the context of microfinance.

Rozina Karmaliani is a professor at AKU SONAM, and in the Community Health Sciences (CHS) Department of Medical College at AKU. She began her career as a community health nurse and preceptor in the CHS department at AKU in 1988. She then went on to complete her MPH in public health administration in 1994, her MScN in public health nursing administration in 1997, and her PhD in nursing in 2000, all from the University of Minnesota. Her areas of research interests include women and children’s health, mental health, women and violence, gender and equity, child abuse and neglect, community health, program evaluation, health systems, nursing education, nursing management, and bioethics.

Cyra Patel (MSPH, MBA) completed her BA in community health at Brown University (2008) and her MSPH in international health at the Johns Hopkins Bloomberg School of Public Health (2012). She has worked in a variety of settings on projects aiming to improve women’s health, from domestic violence to reproductive health. During her time at Brown University, she completed her honors thesis on perceptions of intimate partner violence among South Asian women in Rhode Island and victim service needs. She currently works in the Implementation Research Group at the Sax Institute in Sydney, Australia, where she is working on projects that aim to provide evidence-based health care.

Carla M. Bann completed her PhD in quantitative psychology at the University of North Carolina at Chapel Hill in 1999. She is a fellow of statistics and psychometrics at RTI International and prior to her fellow appointment served as senior director of Program Evaluation and Outcome Measurement at RTI. She has served as the statistician on numerous projects focused on improving maternal and child outcomes.

Elizabeth M. McClure completed her PhD in epidemiology with a focus on perinatal at the University of North Carolina at Chapel Hill (2013). She began her career at the Eunice Kennedy Shriver National Institute of Child Health and Human Development, coordinating a multisite research network to improve neonatal outcomes. Since 2001, she has been at RTI International, conducting research to improve maternal and perinatal outcomes and leading the data coordinating center of a multicountry research network.

Omrana Pasha is an internist and an epidemiologist with more than 18 years of experience working on issues of women’s and children’s health. She completed medical college at the AKU in Pakistan, postgraduate training in internal medicine at Beth Israel Medical Center in New York City, and a MSPH from the University of Alabama at Birmingham. She worked at Save the Children Federation, Inc., in Westport, Connecticut, and the Rollins School of Public Health at Emory University in Atlanta, Georgia, before joining AKU in Karachi, Pakistan, as a member of the faculty in 2004. She currently serves as an associate professor in the Departments of Community Health Sciences and Family Medicine at AKU.

Robert L. Goldenberg completed his medical degree at Duke University (1968) and obstetrics residency at Yale University (1974). He has conducted research on preterm birth and stillbirth, with emphasis on improving outcomes in low-resource settings, resulting in more than 500 peer-reviewed publications. He is currently a research professor at Columbia University (New York, NY) in obstetrics and gynecology.