Exploring the relationship of Domestic violence on Health Seeking behavior and Empowerment of Women in Pakistan

Background: Violence against women is considered as a global problem and the trend for abusing domestically is very much up. This not only violates the human rights but it has got major implications on physical, mental, social and sexual health of a woman. Worldwide little is known on how domestic violence can women’s health and their decision-making power. The present study aimed to explore the relationship between domestic violence on empowerment of women and their health seeking behavior in Pakistan. Methods: This research used secondary data from the Pakistan Demographic and Health Survey data 2012-13. About 3687 ever-married women were selected and interviewed for domestic violence from 248 urban and 252 rural areas of Pakistan. Results: Both chi-square and logistic regression analyses were performed. The results show that women with no education have suffered at lot as a result of domestic violence. Almost 68% women reported that those who have seen their parents beating their mothers have also experiences domestic violence. Logistic regression reveals that women from rural areas are highly likely to be victims of domestic violence (OR=1.28, 95% CI: 1.07-1.52, p<. 001). ANC visit is also found to be statistically significant for those who suffered from domestic violencein Pakistan. Conclusion: Policy makers and other stakeholders should take measures to control and stop violence against women and help them to live dignified human being.


INTRODUCTION
Gender violence is a global problem and domestic violence against women is still a key concern in developing countries [1].Study suggests that violence against women prevails all over the world, though the form may vary from one society and culture to another [2].In December 1993, the United Nations General Assembly adopted the Declaration on the elimination of violence against women [3].About 189 countries government officials accepted gender based violence as a threat to women's health and signed on to the Platform for Action of the 1995 United Nations' Beijing World Conference on Women [4].Previously, little attention is heeded to gender-based violence as a broad social issue, and little efforts were put in place to tackle its underlying causes [5].The most common type of violence that has been reported is intimate partner violence also known as "domestic violence or sexual violence" [6].
The prevalence of domestic violence in Asian countries is very high [7] and the prevalence is higher in rural areas than urban areas [8].There is widespread acceptance of wife beating in many developing countries because of the commonly held norms about gender roles in society.Indeed, the status of women in the family as well as cultural and social factors are responsible for domestic violence on women [7].Aftab and Khan stated that women from lower socioeconomic group have faced a great extent of domestic violence and due to their lower socioeconomic status they cannot leave their relationship [9].According to a report of WHO (2005) factors that put women at risk of domestic violence are individual education, financial autonomy, previous victimization, empowerment and social support [10].According to the ICRW (2002) report men are more likely to beat their wives if they are unable to fulfill their masculinity and have sex with multiple partners to showcase their masculinity [11].Family background is important for children and thus a history of males power abuse to female members in the household suggests that abusive behavior is learned from childhood stages in life [12].
Pakistan is considered as a poorly ranked country on gender development and inequality index and women are considered as a major vulnerable population [13].Women in Pakistan live in a world of strict religious, family and tribal customs that force them to live in fear [14].Pakistani society is mainly male dominated in which women feel inferior to men [15].The dominance by men creates the scope for violence as men try to control their women and women try to resist and thus a violence usually takes place [16].This research is specifically concerned with issues of domestic violence against women and their link with women's empowerment and health status in Pakistan.This is a major issue for Pakistani society where women's low socioeconomic profile and lack of empowerment make them helpless and that has often turned out to be domestic violence and ill health.The aim of this research is to explore the relationship of domestic violence on health seeking behavior and empowerment of women in Pakistan.

Data
The study uses data collected in the Pakistan Demographic Health and Survey 2012-13, a nationally recognized survey conducted by the National Institute of Population Studies (NIPS).The survey was carried out on a total of 498 (248 urban and 252 rural) areas of Pakistan excluding Azad Jammu and Kashmir.Total samples of 14,000 households were selected for interview in which a total of 6,944 households in urban areas and 7,056 households in rural areas were finally selected.The sampling was carried out in two phases.In the first phase, all urban cities and towns were divided into mutually exclusive small areas, known as enumeration blocks and each enumeration block consists of about 200 to 250 households on average and blocks are further grouped into low, middle and high income categories.The urban area sampling frame consists of 26,543 enumeration blocks and in rural areas lists of villages/mouzas/deh were use as the sample frame.In the second stage, 28 households were selected using systematic random sampling technique at each sampling point [17].A total of 12,943 households were successfully interviewed and about 13,558 evermarried women age 15-49 were successfully interviewed.For this research, a total of 3808 women were selected for domestic violence from 13,558 ever married women.Out of 3808 women, 3687 women were finally selected and interviewed for domestic violence.So the sample size used for this particular research was 3687 ever-married women.

Data Collection method and instrument
Pakistan Demographic Health and Survey data 2012-13 used four types of questionnaires-household, woman's, man's and community questionnaire.This research has used household and woman's questionnaire for data analysis.The questionnaires developed by the MEASURE DHS program [17].Close-ended questions were used in the questionnaire.For this research the following topics from the woman's questionnaire were used for data analysis: • Background characteristics of women

Dependent variables
In this questionnaire, a series of questions were used to measure emotional and physical violence.Under emotional violence, the question "Have you ever experienced emotional violence?"has been considered and a dichotomic option ("Yes" and "No") was offered.The respondents who answered, "Yes" were categorized as "experienced emotional violence" and those who answered, "No" were categorized as "did not experience emotional violence".
Exploring the relationship of Domestic violence on Health Seeking behavior and Empowerment of Women in Pakistan Under physical violence, the following questions were selected: • Ever been pushed, shook or had something thrown by husband/partner • Ever been slapped by husband/partner • Ever been punched or hit by something by husband/partner • Ever been kicked or dragged by husband/ partner • Ever been strangled or burnt by husband/partner The response options were "Yes" and "No" and the respondents who answered "Yes" were categorized as "experienced physical violence" and those who answered "No" were categorized as "didn't experience physical violence".
For analysis purpose, both emotional violence and physical violence were merged and one dependent variable was created and named domestic violence.It is a categorical variable with responses options were "Yes" and "No".In our analysis "Yes" stands for "experienced domestic violence" and "No" stands for "didn't experience domestic violence".

Independent Variables
To measure the empowerment of women, the following questions were selected, which represented the independent variables of the study: • Place of residence and the categories are urban and rural • Father used to beat mother and options are yes or no • Empowerment status have two categories like have decision making power and have no decision making power • Beating justified if wife argues with husband and options are yes or no • Occupational status consisted of two categories: currently working and currently not working.• Wealth index and it has got five categories like poorest, poorer, middle, richer and richest • Educational attainment: no education, incomplete primary, complete primary, incomplete secondary, complete secondary and higher • Age of the respondents was divided into four categories such as 10-19 years, 20-29 years, 30-39 years and 40-49 years.For measuring the health seeking behavior of the respondent the following independent variables were selected: • Antenatal care visit consisted of two categories: less than 5 visits and more than 5 visits.• Got STI in the last 12 months with two options: yes or no • Ever had vaccination and response options include yes or no • Pregnancy loses variable was categorized into two options: 0 and more than 1 • Check up after delivery and the options were given: No or Yes • Getting money needed for treatment include options like big problem and not a big problem

Data analysis
The statistical analyses were performed using IBM SPSS.The statistical significance (or alpha) for all analyses was considered at 5 % level.The quantitative data is mainly numerical data and the presentation of the data is done using tables and charts.Firstly, descriptive statistics (frequency distribution) was performed to present the different characteristic of the respondents.Then bivariate association analysis was also performed to determine the relationship between dependent variable domestic violence and other independent variable.In this research, Chi-square analysis has been used.Lastly, logistic regression analysis was used to measure the association between the dependent and independent variables using odds ratio.The reason for using logistic regression analysis is that dependent variable is categorical with category scale Yes" and "No".As explain earlier, "Yes" stands for "experienced domestic violence" and "No" stands for "didn't experience domestic violence".The underlying binary logistic model corresponding to each variable is: Here, P= stands for probability to experience domestic violence 1-P= stands for probability not to experience domestic violence β 0 is the intercept term and β I ' s are logistic regression co-efficients (i=1,2, ...,n) The parameter I refers to the effect of X i on the log odds such that Y= 1, controlling the other X i 's .

RESULTS
The background characteristics of respondents shows that out of 3687 women about 37.5% respondents age were between 30-39 and only 3% women age were between 10-19 (Table 1).About 47 % respondents come from urban areas and remaining 53% are from rural areas.Approximately 55.6% of women had no formal education and only 8.8% of women completed their primary education and around 12.3% of women completed higher education.Almost 50% women have 3-4 children and about 36% women have no children in their family.Almost 11% women confirmed that they were  Table 2 reports about health seeking behavior of the respondents.About 76.5% of respondents indicated that they had less than 5 antenatal care visits and almost 45% of women did not go for health check up after delivery.About 33.6% of women think that getting money needed for treatment is a big problem for them.A vast majority of the respondents (85%) had vaccination before and only 2.2% of women said that they had STI in the last 12 months.About 35.5% of women had more than 1 pregnancy losses.
The figure 1 shows that about 22.2% of women said that they have seen their father beaten their mother in their childhood.
Table 3 presents odds ratio using multivariate logistic regression.The analysis shows that respondents who have seen their father beating their mother have four-time higher chance of getting physically abused by their husbands or partners than comparing to those respondents who have not seen in their childhood.Place of residence plays an important role in domestic violence.This study found that women live in the rural areas are highly likely to be victims of domestic violence.Similarly, as expected the women who are not working are also highly likely to be faced domestic violence compared with the women are employed.Age of the respondents is between 20 and 29 years old only statistically significant.As expected, the women who are between 20 and 29 years old are less likely to experience domestic violence comparing with the women who are teenage (age between 10 and 19 years).This indicates that teenage girls are more likely to be at risk of domestic violence.The odds ratio of educational attainment and wealth index do not show any significant impact on domestic violence in this research.This may be partly explained by the fact that domestic violence in Pakistan is inherent into the deep-rooted cultural phenomenon rather than a socio-economic one.This needs to be further investigate why the country behaves differently than others in the globe.
Table 4 shows the results of associations between domestic violence and health seeking behavior.The analysis reveals that ANC visit is associated almost 5 times less likely to be faced domestic violence.It appears that pregnancy loss is a cause for domestic violence.This study finds that pregnancy loss is statistically significant, which indicates that the women who lost pregnancy more than once are highly likely to be experienced domestic violence.With regards to STI, women who have STI are more likely to be faced domestic violence comparing with the women who had not.Ever had vaccination is not statistically significant.Getting money needed for treatment appears to be significant, which indicates that the households in which money is a big problem for treatment are highly likely to be experienced domestic violence.Check up after delivery is not statistically significant.

DISCUSSION
The study investigates the associations between domestic violence and women's empowerment and women's health seeking behavior in Pakistan.The result showed that how domestic violence affects women's decision making power and their health seeking behavior.The research reveals that ANC visits more than 5 visits; pregnancy loses more than once, women with STI in the last 12-month and households in which money is a big problem for treatment are positively associated with domestic violence.Kabir and Khan (2013) reported that education helps to increase the awareness level for the use of ANC service [18].Women who ever had vaccination are not statistically associated with domestic violence.This shows that women who were abused by physically and emotionally at home were less likely to use up the antenatal care services.This is consistent with other researches in Tanzania, Nepal and in India which shows that domestic violence is associated is less ANC visits during pregnancy time [19,7,20].Result suggests that maternal death rate was high in the United Kingdom due to lack of antenatal care visit [21].
The research findings also indicate that the women who have more than one pregnancy lose are more likely to suffer from domestic violence.It can be explained that domestic violence affects women's physical and emotional health overall and this may affect the overall pregnancy.A study by Cokkinides et al. (1999) found that physical violence during the 12-month pregnancy time is very common and it affects the maternal health [22].One of the noteworthy finding is that women who have got sexually transmitted infections are more likely to be abused by their husbands (or partners) that those who don't have any infections.Research has shown that in some cases pregnancy may act as trigger for violence [6].Another research on violence shows that 47% women experienced violence during their pregnancy and it concludes that domestic violence directly associated pregnancy complications and lack of delivery preparedness [23].
The educational status of women in Pakistan is very low and amongst the lowest in the world [24].The odds ratio shows that educational qualifications of women have no significant association with domestic violence.The bivariate analysis of this research shows that women with no education are likely to suffer more from domestic violence than who have some education or higher education.Both Ali et al, (2011) & Aftab & Khan (2011) reported the same [25,9].The study result shows that about 31% women experienced emotional violence and this is in line with other research findings.One research in Karachi, Pakistan shows that 43% women reported about emotional abuse and about 28% women reported emotional violence in Hyderabad of Pakistan [12,26].About 18% women have shared that they have experienced violence at home.This is in line with a research on married women in Karachi Pakistan and they have found that 27% women have disclosed it [27].
The study presented that women from poorer background are more likely to suffer from domestic violence.This was  supported in a research in Northern India [28].Aftab & Khan (2011) reported that 62.9% of the study population was suffering from under extreme poverty and women from the lower socio-economic group were more vulnerable to violence [9].Findings also indicate that women who between age 20 and 29 years old are likely to experience less violence than the teenagers.A study on Pakistan shows that women's age was not found statistically significant risk factor violence [25].But another research has shown that women aged 20 to 29 years reported significantly less sexual violence as compared with youngest group 15 to 19 years old [23].About 47% women stated that wife beating is justified if the argue with their husbands.This was found most common instigating factor for wife beating [21,16].This shows that women have less decision-making power and have less control of their lives comparing to their male counterparts.Occupational status of women is also an important factor for domestic violence.Women who are not working have greater chance of facing domestic violence.This is probably due to rural women have lack of education and they cannot come out from the socio-cultural barriers.The findings are like to those from another research where they have shown that rural women are less likely to use maternal services because of their low level of empowerment [7].Similar findings were reported by other researches [23,29].The sample size of the study is not very high so it cannot be generalized to all ever married women in Pakistan.Another limitation is that women's reluctance to answer openly on abuse due to emotional bias.

CONCLUSION
The issue of domestic violence in Pakistan has been recognized as a fundamental barrier to women's human right.The prevalence of violence against women in Pakistan is on rise.Domestic violence also affects physical health and mental health of women.Social isolation, lack of knowledge and no control over lives make them unfamiliar with the legal system and keep them away from utilizing health care services.Government, policy-makers and other stakeholders should take immediate measure to design programmes and create legislations to support these poor women who suffer morally, physically, sexually, emotionally and legally.Initiatives should be taken to assist women to realize they have equal rights.

FIGURE 1 .
FIGURE 1. Respondents father ever beat mother

TABLE 1 .
Background characteristics of respondents

TABLE 2
Exploring the relationship of Domestic violence on Health Seeking behavior and Empowerment of Women in Pakistan pregnant during the time of the interview.Table1also shows that about 22.2% of women were working and according to the wealth index almost 18.5% respondents fall under poorest category and 22.7% respondents fall under richest category.

TABLE 3 .
Odds Ratio analysis between domestic violence and empowerment of women

TABLE 4 .
Odds Ratio analysis between domestic violence and Health seeking behavior of respondents