Intimate Partner Violence and Its Associated Factors Among Women in Owo, Ondo-State, Southwest, Nigeria

: Background: Intimate Partner Violence is a public health problem of global magnitude that majorly affects women and is often under-reported. Objectives: This study was carried out to determine the prevalence of Intimate Partner Violence, its pattern, and its associated factors with a view to reducing the burden. Methods: The study was a descriptive cross-sectional study of 347 consenting adult females. Data was collected using the adaptation of the World Health Organization’s Multi-Country Study on Women’s Health and Domestic Violence against Women questionnaire–a cross-culturally validated instrument. Data were analysed using the SPSS version 22 and a p-value < 5% was considered significant. Results: The mean age of the respondents was 41.77 ± 15.64 years. The overall prevalence of IPV was 71.2%. The types of violence in descending order revealed controlling behaviour (49.6%), psychological (47.0%), physical (32.9%), and sexual (19.6%). IPV was significantly associated with marital status (p = 0.023), partners’ being drunk (p = 0.025), money problems (p = 0.002), absence of food at home (p = 0.015), jealousy (p=0.012), sex refusal (p =< 0.001), disobedience to partners (p = 0.003), other situations like children matters (p =< 0.001) and partners’ belief in wife-beating (p = 0.002). Logistic regression revealed partners’ belief in wife beating (OR = 3.734, CI = 1.610 to 8.660, p = 0.002) to be the sole predictor of Intimate Partner Violence. Conclusion: Intimate Partner Violence was prevalent and solely predicted by Partners belief in wife-beating. This, therefore, calls for partners’ education against such beliefs.


Introduction
Intimate partner violence is an important public health problem globally occurring in all settings and among all socio-economic, religious and cultural groups [1].Intimate partner violence (IPV) is a form of domestic violence which includes physical, emotional, sexual abuse, stalking, psychological aggression (including coercive tactics) and controlling behaviours by an intimate partner [1,2].An intimate partner is a person with whom one has a close personal relationship that can be characterized by emotional connectedness, regular contact, ongoing physical contact and/or sexual contact, identity as a couple, familiarity and knowledge about each other's lives [2].Intimate partners include current or former spouses, boyfriends or girlfriends, dating partners, or sexual partners and they may or may not be living together.Women are more likely than men to be injured, sexually assaulted, or murdered by an intimate partner and studies suggest that one in four women is at lifetime risk [3].
The goal 5 of the Sustainable Development Goals as envisioned by the United Nations is to achieve gender equality and empower all women and girls with the elimination of all forms of violence against all women and girls in the public and private spheres adopted as one of its targets [4].Several countries have made efforts in addressing violence against women and girls.
The Council of Europe is the continent's leading human rights organization signed up to the European Convention on preventing and combating violence against women and domestic violence (Istanbul Convention).The Istanbul Convention is an international treaty to tackle serious violation of human rights including violence against women [5].In 2003, African governments committed themselves to ending discrimination and violence against women by adopting a protocol to the African Charter on Human and Peoples' Rights [6].In Nigeria, the federal government adopted a framework and plan of action for the National Gender Policy.The government adopted several legislative and policy instruments including the violence against persons Prohibition Act of 2015, which prohibits female genital mutilation, harmful widowhood practices, harmful traditional practices and all forms of violence against persons in both private and public life [7].
Despite the various efforts to curb or reduce violence against women, the prevalence is still alarming and therefore calls for more collaboration in reducing the incidence of violence against women.
The global lifetime prevalence of intimate partner violence among ever-partnered women is 30% [8].Among the lowand middle-income regions of the world, the World Health Organization found that Africa had the third highest prevalence of 36.6% after South-East Asian and Eastern Mediterranean regions with 37.7% and 37% respectively whereas the high-income regions had an overall estimated prevalence of 23.2% [8].In the United States, about 1.5 million women and 834,700 men experience intimate partner violence annually [3].Furthermore, about 25% of American women reported being targets during their lifetimes and from hospital-based studies, 14 to 35% of adult female patients in emergency departments and 12 to 23% in family medicine offices reported being victims within the previous year [3].A study in Poland found the prevalence of intimate partner violence among female primary care patients to be 35.1% while it was 23.8% among women in England [9,10].A study in Saudi Arabia, reported a prevalence rate of 25.7% and a life time prevalence of 57.7% [11] while in Egypt, a prevalence of 44.1% was reported [12].
In Nigeria, the prevalence of IPV varies across the geopolitical zones.In southwest region, a prevalence of 41.5% among women during pregnancy was reported in Ibadan while 36.7% was reported among women in a primary care setting in Ile-Ife [13,14].However, Ibekwe et al [15] found a lower prevalence of 15.5% in another study in Ibadan.Ajah et al. [16] reported 14.7% and 27.7% among rural and urban women respectively in a community study in South-East Nigeria while 41.9% had been physically abused in a primary care clinic in South-South Nigeria [17].In North-Central Nigeria, a prevalence of 12.6% in current pregnancy and 63.2% previously was found [18].
From the aforementioned, there was no study on IPV in Ondo-state, therefore, this study was carried out to find out the experiences of partner violence among adult female patients in Owo, Ondo-State.The objectives were to determine the prevalence of intimate partner violence, its various forms, the prevailing patterns and the predictors of IPV.The results from this study hope to be used to advocate for the victims of IPV.Moreover, detection might be the first step in breaking the cycle of violence and thus reduce the burden of the problems associated with it, improve the health of the victims and consequently that of the family.

Study Design
This was a descriptive cross-sectional hospital-based study.

Setting
The study was carried out in Owo, Ondo-state, Nigeria.Owo local government area is one of the eighteen local government areas in Ondo State, South-West Nigeria with coordinates 7° 11 ' 46.32" N (Latitude) and 5° 35 ' 12.52" E (Longitude).It links the South-Western part of the country to Abuja, the Federal Capital Territory, of Nigeria.According to the 2006 national census figures, it has a population of 218,886 persons and projected population of 300,000 in 2016 [19].Owo is a home to the Yoruba people but accommodates people of other ethnic origins like the Ebiras, Edos, Igbos and the Hausas.

Study Population
The study participants were recruited from women aged 18 years and above residing in Owo, Ondo-State, Nigeria.

Sampling Strategy
The Inclusion criteria included women aged 18 years and above who gave consent and had or have had intimate partners in the past one year while eligible women who had mental or severe medical illnesses that rendered them incapable of participating in the study were excluded.Three hundred and forty-seven women were recruited in the study.A pilot study was conducted in another centre with the questionnaire administered to thirty-four women to help determine the acceptability and clarity/ambiguity of the instrument items.The administration took between ten and fifteen minutes for each subject and the filled questionnaires were not included in the final study.

Sample Size
The minimum sample size was calculated using the epidemiological formula for the estimation of proportion as follows:

n =
Where: n = the minimum sample size; Z = the standard normal deviate, usually set at 1.96, corresponding to 95% confidence level; p = the proportion of the women with experience of intimate partner violence which is 31.2%(0.312) according to a recent Nigerian study was used [20]; q = the proportion of women with no experience of intimate partner violence: 1-p; d = the degree of accuracy desired which will be taken as ± 5% (0.05 probability).
Therefore, the minimum sample size (n) for this study was estimated as below: = 329.85 An anticipated response rate of 95% was assumed, n = 347

Data Collection
The study was conducted using a pre-tested semistructured, interviewer-administered questionnaire which was also translated into the local language (Yoruba language) for those who did not understand the English language.The questionnaires were administered directly to eligible consenting subjects.
For the purpose of this study, an intimate partner is a person with whom one has a close personal relationship that can be characterized by emotional connectedness, regular contact, ongoing physical contact and/or sexual contact, identity as a couple, familiarity and knowledge about each other's lives which include current or former spouses, boyfriends or girlfriends, dating partners, or sexual partners and they may or may not be living together [2].The questionnaire was adapted from the questionnaire developed by the World Health Organisation (WHO) in a Multi-Country Study on Women's Health and Domestic Violence against Women which has been shown to be crossculturally valid [21].The adjustments on the questionnaire made for this study included questions on the sociodemographic data, husband/partner's attitude to IPV, and subject's financial autonomy.The questionnaire had three sections -A to C. Section A contained the socio-demographic variables such as the subjects and their partners' age, level of education, occupation and social class.Section B was on experience of partner violence: this was categorized into four-partner's controlling behaviour, emotional violence, physical violence, and sexual violence.Screening for Partner Controlling behaviour include asking whether a spouse or partner had ever kept her from seeing her friends, restricted her contact with her family, insisted on knowing where she is at all times, gotten angry if she speaks with other men, accused her of being unfaithful, and whether he controls her access to health care?A 'Yes' response to one or more of these questions listed suggested the presence of partner's controlling behaviour.
Questions on emotional abuse include whether a spouse or partner had ever insulted her or made her feel bad about herself, if the latter had ever humiliated or belittled her in front of others, intimidated or scared her on purpose or had threatened to hurt her or hurt someone she cares about.A 'Yes' response to any of these indicated emotional abuse.
For physical violence, a respondent was asked whether any of her spouse(s) or partner(s) had ever slapped, kicked, dragged or beaten her, choked her on purpose, thrown something at her that could hurt her, threatened her with or actually use a dangerous weapon or object against her and if any of these had occurred in the past twelve months.A 'Yes' response to one or more of these questions indicated physical violence.
For sexual violence, a respondent was asked whether a spouse or a partner had ever physically forced her to have sexual intercourse against her will, whether she had sexual intercourse with the latter because she was afraid of what her partner might do, and whether she has been forced to do something sexual she found degrading or humiliating.And also, if any of these had occurred in the past 12 months.A 'Yes' response indicated sexual violence.Questions on injuries sustained and use of healthcare facilities as a result of IPV were also included in this section.
Section C was on associations of Partner violence.This section focused on possible factors associated with partner violence based on the "ecological approach" used by the WHO Multi-Country Study [8].Questions asked include possible situations that led to the IPV if present, pattern of husband/partner's alcohol use, level of aggression towards other men and his attitude to IPV, family history of IPV on both sides, and financial autonomy of respondents.

Ethical Consideration
Ethical clearance was obtained from the hospital with reference number FMC/OW/380/VOL.XLVII/122.Written informed consent was obtained from all the respondents after explaining the nature of the study to them.Confidentiality was assured by maintaining anonymity on the questionnaires and each completed questionnaire was coded and kept in a secured bag and the data obtained was entered into the computer secured with a password.Each participant was made to know her right to withdraw from the study at any point.

Data Analysis
The data obtained was analyzed using the statistical package for social sciences (SPSS) for windows software version 22 (SPSS 22, Chicago).Descriptive data was presented using tables and charts.Cases of Intimate Partner Violence (IPV) experience was summarised using proportions.Associations between the categorical independent variables and Intimate Partner Violence were assessed with the Chi-square test (χ 2 ).Multivariate regression analysis was done to identify independent predictors of IPV.Level of significance for all the tests was at < 5%.

Results
A total of 347 adult females participated in the study.The respondents were aged 18 years and above with those < 30 years being the majority 111 (32.0%), while the age group 50-59 years had the lowest proportion 49 (14.1%).Mean age of respondents was 41.77 (15.6) years.Married respondents were the highest proportion (54.2%) while those cohabiting were least (2.3%).Most 281 (81.0%) of the respondents were from the Yoruba tribe and Christianity 317 (91.4%) was the dominant religion.The middle class constituted the highest group (37.5%) of the respondents.About two-third (60.5%) earned above the minimum wage (Table 1).About three-quarter 247 (71.2%), had experienced Intimate Partner Violence in one form or the other while the rest of the respondents, 100 (28.8%) had never experienced any form of IPV.Table 2 showed controlling behaviour (49.6%) was the most common type of IPV experienced while sexual violence (19.6%) was the least.
The pattern of physical violence experienced by the respondents showed over one-fifth 75 (21.6%)had been kicked or dragged, 65 (18.7%) had been hit with a fist, 14 (4.3%) had been choked or burnt, while only 2 (0.6%) had ever been threatened with weapons.The relationship between the IPV experienced and the socio-demographic characteristics showed that the middle class 93 (37.6%) had the highest proportion though the difference was not statistically significant (χ 2 = 0.156, p = 0.925).The age group <30 years 88 (35.6%) had the highest proportion that had experienced IPV while the age group 50-59 years 34 (13.8%) had the lowest proportion, but the difference among the age groups was not statistically significant (χ 2 = 0.159, p= 0.057).The association between IPV experienced and marital status was statistically significant (χ 2 = 11.325,p = 0.023).All the respondents who were cohabiting had experienced IPV in one form or the other.The determinants of Intimate Partner Violence among the respondents were assessed by the reported precipitating situations (Table 3).Drunken partner, money problems, lack of food and jealousy were precipitating situations that were determinant of IPV and were statistically significant.Problems with his/her family were not statistically significant determinants of IPV.  4 showed the determinants of Intimate Partner Violence by respondents' and partners' characteristics.The belief by partner that it is acceptable for men to beat their wives was a determinant of IPV and was statistically significant.Difficulties at work, unemployment, and level of financial dependence were not statistically significant.Table 5 showed the multivariate logistic regression of the determinant of IPV.The only significant variable was spousal's belief in the acceptability of partner violence (OR=3.734,CI=1.610 to 8.660, p=0.002).Respondents with spouses or partners that believe that it is acceptable for men to beat their wives were 4 times more likely to experience IPV.

Discussion
Majority of the respondents in this study were below 50 years of age while the highest proportion of the respondents was below 30 years.This age distribution mirrors the Nigerian population pyramid as found by the Nigerian Demographic and Health Survey carried out in 2013, which indicated that Nigeria comprises a relatively young population [20].Married respondents predominated among the participants and this is not surprising because most of the women were within the reproductive age group (15-49 years) which is the period of life when most people marry [21].Monogamy was the most common marriage type found in this study and this probably reflects the influence of Christianity, the religion practiced by most of the respondents which promotes a one-man one-woman marriage philosophy.Majority of the respondents belong to the middle and low socio-economic class which is a reflection of the socioeconomic realities of the country [20].
The overall prevalence of IPV found among the respondents in this study was found to be 71.2%.This was just marginally above the landmark of WHO Multi-Country study reported range of lifetime prevalence between 15% and 71% [20] but was higher than the reported global lifetime prevalence of 30% [8].The similarity with the WHO findings most likely stem from the fact that the WHO study, being a multi-country and cross-cultural study, was carried out partly in African countries which have similar settings to the Among Women In Owo, Ondo-State, Southwest, Nigeria present study area.The higher prevalence found in this study compared to the reported global prevalence might lend credence to the fact that IPV had been under-reported worldwide.
The prevalence of IPV in this study was much higher than that of the average prevalence reported in the developed parts of the world such as Western Pacific region of the world, Europe, and America [8].This might be due to the higher level of awareness, enlightenment, and criminalization of acts constituting IPV in these regions.Study in Saudi Arabia found a much lower prevalence of IPV (11.9%) [20].The difference compared to the finding in this study might be due to the methodology used in this study.
The finding in this study was also higher than the average WHO-reported prevalence of 36.6% in Africa [8].This could be as a result of the varied settings and population studied.Compared to findings among ever-married women in the Nigeria Demographic and Health Survey (NDHS) 2013 [20], the overall prevalence in this study was almost three-times higher.The reason for this difference might be because of the timing of the studies and could also be due to the larger sample size.
The most common type of IPV found in this study was controlling behaviour which was experienced by almost half of the respondents.This was similar to that seen in the WHO Multi-Country study where controlling behaviour was the most common IPV experienced [21].Controlling behaviour suggests male control over their spouses'/partners' behaviour is normative to a large extent in the study environment as in many other parts of the world [21].In most relationships in African setting, the male partner is usually older in age and therefore tends to control the female and the relationship dynamics.This was further supported by a study in Spain where controlling behaviour was more frequently reported among couples where the man was older than the woman [22].Furthermore, controlling behaviour has been found to be closely associated with other forms of violence [21,22].
Physical violence ranked third most common IPV in this study, having been experienced by one out of every three respondents in varying severity.It is the major cause of injuries and mortality associated with IPV worldwide and it was what turned the attention of the world and other relevant organizations to IPV [1-3, 8, 21].This finding falls within the reported ranges found in the WHO Multi-Country study [21].However, it was higher than studies reported in Slovenia [23], the NDHS [20], and Nigeria [18].The findings from the present study were lower than studies in Port Harcourt, Nigeria and Guinea [17,24].
Sexual violence was the least common type of IPV among the respondents in this study.This finding was largely consistent with those of other studies [17,20,21].The prevalence in this study was almost three times higher than that reported by the NDHS [20].It was two times more than that reported in Jos, Nigeria [18].It falls within the range of the WHO Multi-Country study findings [21] but was lower than the finding in Guinea.[24] The variation in the prevalence of sexual IPV could be dependent on the ability and willingness of the participants to divulge sexual issues which are mostly intimate matters and many of the participants might not want to divulge information about such issues.
This study found that respondents less than 30 years of age had the highest proportion of IPV experience.This was similar to findings in other studies, where IPV was commoner among the younger women [20,21].This might suggest that violence starts early in relationships and that younger men may exhibit more IPV tendencies than older men.It might also suggest the older women might have a recall problem or bias or had developed coping strategies to this problem and so are less likely than the younger respondents to report it [21].Marital status was found to be associated with IPV in this study.The highest proportion of IPV was found among the separated/divorced respondents.This finding was similar to previous studies [20,21].Study in Kano, Northern-Nigeria reported a similar finding that marital status was associated with IPV and being married reduced the risk of IPV while divorced women were more likely to have experienced IPV [25].Intimate partner violence might have been the primary cause of the marital disharmony/divorce.Also, these group of respondents were probably more willing to disclose their IPV experiences since they were no longer in the abusive relationship compared to those who were currently in a married relationship.Respondents who were currently married might not be willing to disclose their IPV experience in order to protect their relationships.
In this present study, the factors precipitating IPV were partner's alcohol use, money problems, absence of food at home, spousal's/partners' jealousy, respondents' refusal of sex, and respondents' disobedience to spouses.Other precipitating factors were not taking adequate care of children, suspicion of respondents by their partners, and issues relating to other women.Capaldi et al [26] reported men's jealousy was associated with male to female partner violence.A study in Botswana reported that disobedience to spouse and refusal of sex had an association with IPV [27].Similarly, several studies [25,27] reported refusal of sex as a factor precipitating IPV.This might show that many partners do not want the females to exercise their rights to negotiating sex.It is said that in the absence of legitimate means of displaying masculine success, and to deal with feelings of disempowerment, the dominant cultural model of ideal masculinity finds its expression in male performances that dominate women and celebrate aggressive male sexual behaviour [28].
Several studies [27,29] also accounted respondents' disobedience to partners as a cause of IPV which further buttress the issue of male dominance in many relationships worldwide.The other factors precipitating IPV like food, money issues, not taking adequate care of children, and partner's suspicion of infidelity were also reported in other studies [27,29].Alcohol use was also reported in several studies [20,27,30] as a factor precipitating IPV.Alcohol is a known cause of many vices [30].
The predictor of IPV in this study was spousal/partner's acceptance of IPV, in fact, the odds of IPV was about four times higher in respondents with partners that believed that it was acceptable for men to beat their wives than those respondents whose partners did not share this belief.This was in consonant with findings from other study [25] where males' tolerant attitudes to IPV was associated with spousal IPV.The acceptance of IPV might be connected to the existence of some cultural norms such as beating of wives in some developing countries like Nigeria may be seen as a form of discipline rather than being a violent behaviour.It might also function as a means of enforcing conformity with the traditional role of women in such societies [17,21].Surprisingly, studies have shown that even women in some cultures justify wife beating under certain circumstances [20,21].

Conclusion and Limitations
In conclusion, Intimate Partner Violence among adult females in Owo, Ondo-state was highly prevalent and controlling behaviour was the most common form of IPV found among the respondents while sexual violence was the least type experienced.Surprisingly, spouses still believe that it is acceptable for men to beat their wives and this was a predictor of IPV.Women who had such spouses/partners are four times more likely to experience IPV than other respondents.This therefore, calls for education of the spouses and the community against such belief and culture.
The limitation of this study include the effects of recall and report bias on the part of the respondents.Also, the design of this study did not include male partner to confirm the belief of battering declared by the female partner and their thought on what triggers the violence in them.Despite these limitations, this study provided evidence of high prevalence of IPV among adult females in Owo, Ondo-state.

Table 1 .
Socio-demographic characteristics of the respondents.

Table 2 .
Types of Intimate Partner Violence experienced by respondents.

Table 3 .
Determinants of IPV by precipitating situations.

Table 4 .
Determinants of IPV by respondents' and partners' characteristics.

Table 5 .
The predictors of IPV among the respondents.