Intimate partner violence against Iranian women

Introduction

Violence against women is a worldwide phenomenon. It is of public health importance and related to violation of women′s rights. It affects millions of women around the world ^[1],[2] and occurs in various cultures, ethnicities, nationalities, religions, socio- economic classes and literacy levels. ^[1],[3] The United Nations has defined violence against women as any act in the form of physical, sexual or psychological harm or suffering to women in public or private life. ^[4] It is classified into two parts--non-intimate or domestic violence (DV) and intimate partner violence (IPV). ^[5] DV is the most common form of violence against women globally, ^[6] and 1-3 of 5 women will experience a form of violence in their lifetime. ^[7] According to the WHO report in 2013, almost a third (30%) of women in the world experienced DV, and in some areas it reached 38%. ^[5] In Iran, the overall prevalence of DV against women is estimated to be 48.9%. ^[8] According to Sahraian et al., social violence is more than 80%; 73% of wives were regularly beaten by their husbands, and the rate of humiliation was estimated to be 77.2%. ^[9] Based on a review article in Iran, the prevalence of DV varies based on the type of violence and is higher than 90% in types of physical, psychological and sexual assault. Moreover, 81.6% of DV victims encounter economic violence. A study by Kazemain et al.10 found that physical and psychological violence was experienced by 93% and 88% of women, respectively.

Nowadays, IPV in the community is a major obstacle to achieving Social Development Goals due to gender inequality. ^[11] IPV is a risk factor that has an adverse effect on various aspects of women′s health, both directly and indirectly. ^{[5],[12],[13]} In terms of sexual and reproductive health, it can increase the likelihood of problems including sexually transmitted diseases, pelvic inflammatory diseases, menopause, unwanted pregnancy, low birth weight, preterm labour, stillbirth and neonatal and perinatal death. ^{[5],[13],[14],[15],[16],[17],[18]} IPV can lead to an increased risk of physical and mental illness as well as an increase in the tendency for risky behavioural disorders such as injury, suicide, suicidal attempts, substance abuse, drug abuse, cigarette and alcohol consumption and murders. ^{[5],[13],[14],[16],[19],[20]} DV is often not reported or under-reported because of social norms, stigma, taboo, fear, shame, revenge and its high sensitivity. Therefore, accurate data are difficult to obtain. ^[13],[20]

Women in developing countries are more affected by DV than those in developed countries. ^[21] The prevalence of IPV was highest in the WHO African, Eastern Mediterranean and South-East Asia Regions. ^[5] Iran as a developing country is located in the eastern Mediterranean region of WHO where there are little reliable data on IPV. ^[14],[22] Based on the latest studies, it was found that family problems, belonging to a patriarchic family, accepting the act of violence and the types of reactions towards husband′s violence were factors related to IPV. In addition, spousal addiction and the wife′s job were also related to experiencing IPV besides the educational level. ^{[22],[23],[24],[25]} An educated woman working as a nursing staff is expected to face less violence than a home-maker due to financial independence and higher socioeconomic status. ^{[23],[24],[26]} Therefore, due to the high prevalence of IPV in Iran, we aimed to (i) estimate the prevalence of DV in Iranian women--home- makers and those employed (nurses); and (ii) compared the prevalence of IPV in women with husbands with and without history of substance use.

Methods

This cross-sectional study was done on 240 married women in June 2014 in two phases to evaluate the prevalence of DV. The study setting was Qom, a metropolis in central Iran. ^[27] The first phase was done on a representative sample of married women who lived with their husbands as home-makers. In the second phase, the prevalence of DV was assessed in a random sample of women who were working outside their homes as nurses. The prevalence of violence in home-makers was compared with those working as nurses. One hundred and twenty home-makers and 120 married women nurses were enrolled in the study. The study groups were matched with respect to their ages and their husbands′ ages within 2 years. All were residents of the district of Qom.

The first group included a representative sample of home- makers and the second group was representative of all married employed women (nurses), who were selected by the multistage random sampling method. For selecting the sample, first the urban areas of Qom were divided into four parts: north, south, east and west (geographical divisions) based on stratified sampling. Later, from each cluster, a healthcare centre was chosen randomly and the participants were selected on the basis of systematic sampling in that centre according to the household identification number. Primary healthcare was delivered to all women and their children free of charge and access to these cares had a high coverage. ^[28],[29] A random sample of home-makers referred to healthcare centres were considered as the first group. Married women working as nurses in two public and private hospitals of Qom were selected as the second group.

Being married and living with husbands in Qom were the inclusion criteria. Those who were unwilling to participate in the study were excluded. In addition, the prevalence of DV was estimated according to the participant′s job and husband′s opium use. This work was financially supported by Qom Welfare Organization, and the ethical committee of Qom University of Medical Sciences approved the study protocol. Verbal informed consent was taken from all participants in the study.

Data collection was done using a proforma and consisted of demographic questions and questions about the causes of violence beside the modified questionnaire of DV CTS-2, which was made by Strauss in 1996. The CTS-2 questionnaire included multiple- choice questions with Likert-scale format (never, low, medium and high). ^[30],[31] The CTS-2 questionnaire could measure different types of violence in women and men including physical, psychological, verbal, economic, sexual and injury-type of violence. The economic subscale was added to our revision of CTS-2 questionnaire and we used the women CTS-2 questionnaire. The modified version of women CTS-2 questionnaire includes 43 questions about different types of violence: physical (12 questions), psychological (10 questions), verbal (6 questions), economic (4 questions), sexual (5 questions) and injury (6 questions). The mean scores of each sub-scale of the questionnaire were calculated based on the woman′s experience of violence in the most recent years (yes 1 and no 0). The prevalence of DV is estimated based on the yes responses of participants in each subscale of the questionnaire. The reliability of this questionnaire was calculated by Cronbach′s alpha coefficient for each sub-scale of the questionnaire and varied from 0.79 to 0.95. In addition, the content validity of the questionnaire has been evaluated and approved by experts in psychology, epidemiology and biostatistics. It has also been used in other Iranian studies as a reliable tool. ^[32],[33]

Data collection was done by a self-reported questionnaire that was distributed among the participants in both study groups. The data were analysed using SPSS software. For data analysis, Spearman′s correlation coefficient, chi-square and t test were used. The level of significance was considered to be 0.05. The most important factors affecting physical violence among employed participants and home-makers were evaluated by multivariate binary logistic regression. Fitness of model was determined using the Hosmer-Lemeshow statistic and R-squared measure of goodness of fit (R2) and Akaike information criterion.

Results

The mean (SD) age of the participants was 33.1 (4.5) years and of their husbands was 37.8 (4.7) years. The mean (SD) duration of the participants′ marriage was 12.6 (9) years [Table - 1]. The mean (SD) parity was 1.7 (1.35). Overall, 43.8% of participants and 47% of their husbands had an academic background while 6.7% of husbands were using narcotic drugs. Consanguineous marriages occurred in 32% of participants. Financial problems were reported by 53.3% of participants and 26.7% of participants were living in rented homes.

Table 1: Demographic characteristics and the prevalence of different dimensions of violence in the participants

Nearly all (98.8%) the participants were exposed to at least one type of violence. In addition, the prevalence of verbal and psychological violence was 95.4% and 80.8%, respectively. Moreover, physical violence occurred in 28.8% of participants, economic violence in 45%, sexual violence in 28.3% and injury- type violence in 14.6% of participants [Table - 2]. There was no significant difference among employed women and home-makers (p>0.05). The mean scores of verbal, psychological and physical violence corresponded with the most frequent types of violence in the two groups [Table - 2]. There was no significant difference between the mean scores of violence in all dimensions except in economic violence (p>0.05). The mean of economic violence in employed participants was approximately half that in home- makers and this difference was statistically significant (p=0.02). In addition, the higher prevalence of violence in different dimensions was observed in home-makers than in employed participants except in injuries that were higher in employed participants. Nevertheless, no significant difference was observed between the two groups (p>0.05). Physical, psychological violence and injury were significantly higher in participants whose spouses were addicted (p<0.05; [Table - 3]). There was no significant difference between participants with addicted and non-addicted husbands with regard to verbal, economic, sexual and overall violence.

Table 2: Comparison of the mean (SD) score and the prevalence of violence and its dimensions among employed and home-makers in the first phase

Table 3: Comparison of the prevalence of violence and the spouse's use of opium

A multivariate regression model was used to assess the predictive factors of physical violence against women [Table - 4]. It showed that financial hardship, a history of divorce in the woman′s family and spousal addiction were the most important factors for physical violence. These results were achieved after control of some variables such as educational levels, age of spouse, length of marriage, accommodation or remarriage. Spousal addiction had the highest impact on physical violence, increasing it four-fold. In addition, history of divorce in the woman′s family and financial problems increased the chances of violence by three- and two-fold, respectively (p<0.05).

Table 4: Factors affecting physical violence in employees and home-makers based on logistic regression model

Discussion

Our study shows that nearly all the participants (98.8%) had experienced at least one type of violence in Qom, Iran. The results of another study in married women visiting Ahvaz′s health centres, Iran, showed that prevalence of violence against women varied between 27% and 83%. ^[34] However, a study from India using a standard violence questionnaire showed a higher prevalence of physical and sexual violence (78%) than emotional violence (1.8%).7 Another large population-based cross-sectional survey in married women residing in urban Rasht (northern Iran) showed that 57.1% suffered psychological aggression, 27.6% physical abuse, 26.6% sexual abuse and 6.9% injury. ^[26] Another systematic review and meta-analysis in Iran showed that the prevalence of IPV was 66%. The geographical distribution of this study showed that the prevalence of DV was 70% in the east of Iran, 70% in the south, 75% in the west, 62% in the north and 59% in the central part of Iran. ^[35]

According to our study, verbal and psychological violence were the most common types of violence. Since these types of violence are more subjective, these reports could be based on the perception of the woman and her expectation from life, knowing her rights and little knowledge about life skills. ^{[8],[23],[24]} However, the estimated prevalence of violence in physical (28.8%), economic (45%), sexual (28.3%) and injury types (14.6%) was similar to the results of other studies in different cultures. ^{[25],[36],[37],[38],[39],[40]} Other studies showed the overall prevalence of DV against Iranian women to be 48.9%.8 In a study by Sahraian et al., the prevalence of social violence was more than 80% and for humiliation, it was estimated to be 77.2%.9 Moreover, 81.6% of victims of DV encounter economic violence. Kazemain et al. reported that the prevalence of physical and psychological violence was 93% and 88%, respectively. In another study on infertile women in Iran, the prevalence of DV was estimated to be 61.8%. ^[25] It has been suggested that unintended pregnancy may also be related to DV. ^[17] However, other studies both outside Iran7, 16, 38 as well as those conducted in other areas of Iran have reported lower prevalence of DV in comparison with our study. ^{[40],[41],[42],[43]} The discrepancies observed among results of various studies might be attributed to the religious and cultural context of the city of Qom, the mood and culture of men against women and a conservative approach of men towards women. Although Islam is opposed to any form of violence, in a study by Mahapatro et al. in India, it was shown that Muslim women are at greater risk of physical, psychological and sexual violence than women who were Buddhist, Christian or Hindu.44 However, the effect of religion on DV against women requires further investigation.

The prevalence of physical and sexual violence in our study was 28.8% and 28.3%, respectively, which was about 50% lower than that of psychological violence. Three different studies in Pakistan, Iran and Northwestern Ethiopia ^[38],[42] indicated higher physical and sexual violence than our study. The lower prevalence of sexual and physical violence in our study may be due to under- reporting by women because of fear or shame. In addition, because beating women is a crime in Iran, violence in verbal, psychological and financial forms tends to increase because men fear facing prosecution for abusing their wives. Our study showed that there was no significant difference between home-makers and professional nurses regarding different types of DV. Ardabily et al. showed similar results in Tehran, Iran.25 Only the mean score of economic violence in employed participants was significantly lower than that of home-makers. Therefore, women′s job or financial support can be an effective protective factor against violence. Other studies showed that participants′ job was associated with violence and participants who were unemployed experienced more violence than those employed in government or those who were self-employed.1, 39, 40 This may be due to unemployed women′s financial dependence on their spouses. Another study in 2012 in Jeddah, Saudi Arabia, showed that unemployed participants who were financially dependent were at nearly 1.5-fold risk of DV.^[16]

We found that financial problems, a history of divorce in family of women and spousal addiction were important factors for DV. Shorey et al. found significant relationship between addiction of the husband to alcohol and drugs and DV.45 The effect of spousal addiction to cigarettes, alcohol or narcotic drugs on DV has been reported in several studies.16, 39, 45-47 Another study from South Africa^[48] showed that alcohol consumption, aggravated by the other partner and lack of family support were factors contributing to violence. On the other hand, good economic situation and emotional support by wife′s family were protective factors for DV against women. Similar to our study, some studies report a significant relationship between family income and DV.^{[38],[43],[47]}

Although we could estimate the prevalence of DV and its various dimensions, it seems that due to some problems in marital and sociological dimensions, the prevalence of violence has been underestimated.24, 25, 34 According to the society′s cultural structure and the acknowledged helplessness in Iranian women and mothers′ sense of forgiveness, many cases of violence against women, particularly in discussions of emotional, psychological, financial and especially sexual violence could not be reported.14, 34, 39 However, to obtain valid data, we tried to establish trust and a friendly relationship with women. One of the limitations of our study was the fear of mothers in revealing their responses and fear of further violence by their husbands, a concern that we tried our best to dispel by assuring them that the information would remain confidential. Future studies should study the effect of religious, legal and cultural differences in different areas, especially on high-risk groups of women. Life skills training, especially problem- solving skills, anger control and effective relationships in men are necessary to reduce violence. Moreover, assessment of the impact of DV on children′s performance and future perception could be evaluated in longitudinal studies.

Conclusion

The prevalence of DV against women in the city of Qom seems to be high. Women′s job was not related to violence prevalence except in economic violence. Nevertheless, spousal addiction was related to higher violence against women in the form of physical, psychological and injury violence. Poverty, addiction and divorce were the three most important predisposing social factors of violence against women. Due to the existence of positive correlation between mental health and problem-solving skills, anger control and dealing with emotions, life skills training can be an effective factor in reducing violence in the family. Therefore, development of life skills education in all groups of society such as women, young people and mothers should be essential.

Acknowledgements

We are grateful to the Research Vice-chancellor of Qom University of Medical Sciences as well as to all the women who participated in the study. This study was financially supported by Qom Welfare Organization.

Conflicts of interest. None declared