Rural and urban prevalence of sexual assault against women in an African population

: Background : Sexual assault is about the most dehumanizing form of gender based violence against women worldwide. Nigeria and many other countries in Africa do not have National data on women sexual assault. This survey is aimed at generating data on sexual assault against women in Osun state, Nigeria. Objective: To determine the patterns of sexual assault against women in Urban and Rural areas of Osun State in South western Nigeria. Methodology: A cross section survey using interviewer administered questionnaire was carried out among selected 1,200 women aged 15 years and above in urban and rural areas, between August and December 2014. The questionnaire was patterned after WHO Multi-country study on women’s health and domestic violence data instrument. The data were analyzed using SPSS version 17.0. Results: Mean age of the respondents was 23.75 ±(13.22) years in rural area, in the urban area it was 27.69 ±(10.23) years. 46 % and 54 % were married in urban and rural areas respectively. The prevalence of completed rape was 10.0 % in urban and 9.2 % in rural, while that of attempted rape was 31.4 % and 20.0 % in the urban and rural area respectively. Women in the rural areas experienced repeat sexual assault and suffered non-genital injuries more the in urban area. Having partner and living in urban area were associated with female sexual assault. Conclusions: Sexual assault against women constitutes public health issue in Osun state with rural women incurring greater negative health consequences. Primary prevention strategies should focus at young men and women in both rural and urban areas of the state.


Introduction:
All over the world gender based sexual assault against women is a recognized problem with negative consequences on the health of women. 1,2,3 . Rape is probably the most brutal of all forms of gender based violence against women. Rape is defined as physically forced or coercive penetration of the vulva, including virginal and or anus. 4 Sexual assault comprised of a range of offensive actions including: sex without consent (rape) sexual control of reproductive rights, other sexual manipulation as Forced kissing, breast and genital fondling, and attempted rape 5 . Rape and other forms of sexual violence is a violation of the fundamental human rights of the victims as contained in the United Nation general assembly declaration on the elimination of all forms of gender based violence against women 6 . In southern part of Nigeria (the study site) the legal code that provided for sexual offence is under Section 357 of the Criminal Code which states that: "Any person who has unlawful carnal knowledge of a woman or girl, without her consent or with her consent, if the consent is obtained by force or by means of threats or intimidation of any kind, or by fear of harm, or by means of false and fraudulent representation as to the nature of the act, or in the case of a married woman, by personating her husband is guilty of an offence which is called rape." 7 The above legal provision connotes societal disapproval of the act of sexual assault and this is routed in the culture, tradition and religious belief of the people of Nigeria and many other countries. Despite such rejection, rape and other forms of sexual assault are being committed frequently in Nigeria and other Nations. 4,8. . The true prevalence of rape and other sexual assaults in the population may never be known due to gross under reporting by the victims 8 . For every one act of rape reported eight others had occurred with the survivors keeping mute 9 . The reasons for under reporting included stigmatization of victims rather than the perpetrators, self blame by the victims and low conviction rate aside from time consuming judicial process 10 . Prevalence of rape varies from one region to another depending on the type of study (population or institutional based). In the same region, differential prevalence have been observed between rural and urban settlements 1,10 . Prevalence ranging from 5.6% to 50.7% had been reported 4,11,12 . Age and low socio-economic status are among factors that influences risk of a woman been raped, the young in their teens are at increased risk 10,13 . In the United State of America (USA) research indicates that women of color especially black have elevated risk of sexual assault 14 . However, all over the world every woman or girl is at risk of sexual assault. The rampant occurrence of female sexual assault was captured in a study by Fattah and Kabir titled "no place is safe sexual abuse of the children in rural Bangladesh" 15 . A woman who experiences rape is most likely to suffer physical violence and emotional abuse and there is repeat perpetration of all of these violent acts over time. There is association between sexual violence and physical as well as emotional torture 16 . Victims of rape and other sexual violence are exposed to serious health hazards including physical injuries, sexually transmitted infections such as Gonorrhea, syphilis and HIV/AIDS. For many women psychological effects of rape may be profound and it includes loss of self-esteem, anxiety, alcoholism, depression leading to post traumatic stress disorder PTSD 17 . As part of measures to tackle the menace of sexual violence against women WHO recommended systematic data collection about sexual assault. In United state of America there is biennial sexual and criminal violence survey and a United Nation multicountry survey on sexual violence was carried out in Latin America and China in 2010 8 . Nigeria has no National data on female sexual assault. There is a need to bridge this gap. This survey was aimed at generating regional data as well as a follow up to a report of audit of cases of rape treated at general hospital in Osogbo Osun State over a seven years period (2004 to 2010). Urban and rural areas were surveyed with a view to bringing out the differences that may be present between the two places. Urban centers are characterized by overcrowded housing and living conditions, availability and usage of social media, loss of cultural and traditional control with increased person to person interaction. These features are likely to have effects on sexual behaviour. Also, few studies have reported on urban and rural patterns of female sexual assault especially in Nigeria. Methodolgy: Study area: Osun State is located in south western Nigeria and has a population estimate of about four million in 2013 National demography and health survey 18 . There are thirty local government areas (LGA) and one area office (EA) in the state. The communities are Yoruba speaking, and majority of the inhabitants are traders, artisans, farmers or civil servants. A cross sectional survey using questionnaire was carried out in the state from 1 st August to 30 th December 2014. Multistage sampling technique was used. In stage one, out of thirty LGAs in the state, twelve LGAs (40%) were randomly selected using ballot paper technique. The LGA was used as the sample frame for the study. Using Leslie Fischer's formular for calculation of sample size when the population is greater than 10,000. With rape prevalence of about 6 %, reported by Daru et al 2011 4 a sample size of 87 was obtained per LGA and a total of 1,044 for the 12 LGAs. We projected questionnaire attrition (incomplete and inappropriately filled questionnaires) rate of 20% which is 209. Thus a total of 1,253 questionnaires were calculated for the state. Stage two, cities, towns and villages in the twelve selected LGAs were stratified into urban and rural areas, using the state administrative divisions. One area each (site) from urban and rural areas were randomly selected using ballot method from the lists of all urban and rural areas in each LGA. In all, twelve urban and twelve rural areas were selected. In stage three each selected site (urban and rural) was broken down into Enumeration areas (EA) according to National population Commission 2008 document. Two enumeration areas were randomly selected in each urban and rural area making a total of forty eight (48) enumeration areas made up of 24 EA in urban and 24 EA in rural areas In stage four, two streets were randomly selected in each EA and the houses in the selected streets were numbered and every Fourth house (systematic sampling) was selected for the survey. All females aged 15 years and above in the selected houses were interviewed individually using trained female (same gender) interviewer. This continued until the questionnaires allotted to the EA were exhausted. (Minimum age of 15 years was chosen in conformity with WHO-multi country study 2005 1 ). Urban areas have higher population density per EA compared to rural areas 18 . Questionnaires were therefore administered in ratio 3:2 approximate 32:21, in urban and rural EAs respectively. A total of 768 and 504 questionnaires were administered in urban and rural areas respectively, giving a total of 1,272 questionnaires administered in the state. This was higher than 1,253 calculated. However, the increment can only improve rather than detract from the study. Data Instrument: The questionnaire was patterned after WHO Multi-country data instrument, while we adopted Centre for disease control CDC classification of rape and sexual assault (2014) 19 with modification. It has both open and closed ended sections. Information sought included socio-demography, whether or not the respondent had experienced any or all of the three types of sexual assault covered in this work (Type 1. unwanted sexual contacts such as fondle beast, buttock, peck and kiss. Type 2, attempted rape included actions like pulling down a woman, removing her cloths and under-wears or panties and trying to lie on top of her. Type 3, completed rape means a man penetrating a woman's vulva, vagina or anus with his penis or any other part of his body especially fingers without her consent.) Was the perpetrator a partner or non-partner, did she sustained injury to the genital or other parts of the body and if she ever reported to the police. We sought the opinion of the respondents on why men sexually assault women and the ways to prevent the assault. In the questionnaire, we described the actions that constituted each of the three types of sexual assault to ensure the respondents understand the object of the questions. Where an eligible person was not in the house at the time of first visit a repeat visit was made. Verbal consent was obtained before the questionnaire administration. Souvenir a piece of tablet bathing soap was given to each respondent at the completion of questionnaire administration. The data was entered into computer using SPSS 17.0 version. Validity of data entered was ensured using double entry and random manual checks. Frequency tables and charts were generated. Bi-variate analysis was done using Chi-squared test, while logistic regression analysis was done and generating Odds ratios and 95% confidence interval. Significance level was considered at p values <0.05 Exclusion criteria: Females aged less than 15 years, resident period less than six months as at the time of the survey and those who did not consent to the study. Ethics: The ethic committee college of health sciences, Osun State University gave approval for the study. RESULTS: Out of a total of 1272 questionnaires administered, 1,200 (94.3 %) made up of 720 in the urban and 480 in the rural areas were properly filled and returned and were analyzed. Table I, Showed the socio-demographic features of the respondents. The average age of the respondents were 23.75 ±(13.22) years and 27.69 ±(10.23) years in the rural and urban areas respectively. In the urban area age group 15 to 24 and 25 to 34 accounted for 48.0% and 26.1% respectively. In the rural area respondents in the same age groups represented 40.5% and 30.3% respectively. Educational status of the respondents showed that 30.4% and 3.1% had tertiary and nil formal education in urban area respectively. In rural area the respective figures were 23.8% and 7.5%. The marital status of the respondents showed that 42.1% and 56.0% were married in the urban and rural settlements respectively.   Table iii showed different categories of sexual assaults in both urban and rural areas and the perpetrators responsible. In unwanted sexual contact male partners of the victims were responsible in 55.3 % and 67.4 % in Urban and rural area respectively. In 53.2 % of attempted rape and 70.0 % of completed rape male partners of the victim were the perpetrators in urban centre, in the rural area the respective figures were 57.1 % and 52.3 %. Male Partners commit more sexual assaults in both urban and rural areas, but it was only in the latter that it was statistically significant. In   Survivors in both urban and rural areas had nongenital injury and it was statistically significant. In figure 5: The prevalence of genital injuries in attempted rape and completed rape in the rural area were 19.0 % and 29.5 % respectively, while in the urban area the figures were 8.9 % and 30.6 % respectively. Survivors in both urban and rural areas had anogenital injury from sexual assault.  Predictors of female sexual assault includes living in urban settlement, being sexually active, and having a partner.

Discussions:
The average age in both urban and rural areas were similar, being 23 ±10 years and 27 ±9 years respectively, which showed that majority of the respondents are women within the reproductive age and who constitute a significant target group of sexual assaults 19 .
School attendance level is quite high in this study as only about 3 % and 7 % in the urban and rural areas did not attend school and it may be due to the relatively young age group that constituted significant component of the respondents. This finding is in agreement with 2013 National demography and health survey NDHS as reported by National population commission (NPC) of Nigeria 18 .
The average prevalence of completed rape in the urban area was 10.0 % which is not far from 9.2 % recorded in the rural area. The prevalence in this study is higher than rape prevalence of 5.7 % and 5.0 % reported respectively by Olaleye and Ajuwon 11 21 . In this survey we reported attempted rape (a non-penetrative sexual assault) prevalence of 26.2 % and 31.3 % in the rural and urban areas respectively; this is similar to nonpenetrative sexual assaults prevalence of 14 and 43.9 percent documented by these authors in the same studies cited above. The prevalence of unwanted sexual contacts (another type of non-penetrative sexual assault) in this study was 64.6 % and 58.8 % in the rural and urban areas respectively. This work differs from the older studies in two respects; we reported rural and urban patterns because of differences in physical and social context between the two areas and its possible implication on sexual behaviors. Also, we re-defined nonpenetrative sexual assault into two distinctive groups based on the different levels of violence involved. These prevalence figures should be seen as only tip of ice berg due to gross under reporting of sexual violence against women 8,10 .
In this study however, we have addressed the problem of under reporting by ensuring confidentiality and the use of same gender interviewer with flexibility that allowed self filling of questionnaires by those respondents that requested such. The prevalence of unwanted sexual contact was higher in rural area than the urban, 64.6 versus 58.8 percent, while for attempted and completed rape the prevalence was marginally higher in urban area compared to rural area. The reason for higher prevalence of the attempted and completed rape (which are more severe violence) in the urban may be due to higher population pressure, higher number of unmarried older men and women as reported by Thomas,Scott & Esquibel 22 . Other possible explanation could be closer social economic interactions and other confounding variables of the urban living condition. However, our work did not cover this aspect. Across rural and urban areas there is high prevalence of unwanted sexual contacts in excess of attempted and completed rape, this may be partly due to socio cultural factor in Yoruba land, that permits an acquaintance male to torch or fondle sexually sensitive parts (such as the breast and buttox) of an unmarried woman. This is referred to as "Ta ge" or "O ge tita" in Yoruba culture. This act is a form of sexual assault or harassment as consent is implied but not obtained, intervention programme should recommend abrogation of such practices. In this study, rural women suffered more repeat unwanted sexual contacts and attempted rape than urban women. Renninson, Dekeseredy & dragiewicz 23 reported similar findings in a study titled intimaterelationship status variation in violence against women urban, sub-urban and rural differences. However, repeat completed rape is similar in both areas about 1 in 9. This is in agreement with Fattah and Kabir report from Bangladesh 15 . The overall prevalence of re-victimization of sexual assault reported in this study is in agreement with 23 % re-victimization rate reported by Walsh, DiLillo, & Messman-Moore (2012) 24 . This finding raised the question, what factor or factors make a woman to be at risk of repeat sexual assaults? In the same study Kate Walsh et al 2012 observed that re-victimization is commoner in women with emotional dysregulation and poor risk perception, analysis of these factors were beyond the scope of our study. Never the less, this study showed that residing in urban area and having a male partner ( boy friend or man friend) were significantly associated with attempted rape, p value <0.05. Similarly, having male partner was independently associated with sexual assaults p value <0.05. Intervention programme need to recognize and address the context and content boy friend and man friend relationship as it affect sexual assault against women. Being sexually active and age of sexual debut (which was nineteen years in this survey) were not associated with increased risk for sexual assault. This is not surprising because being sexually active and age of initiation were determined by individual woman voluntarily in most cases, where as the perpetrator (external factor to the victim) was solely responsible for sexual assault.
There was no female perpetrator reported in this survey. This may reflect the fact that same sex sexual offence is not common in Osun State. Partners were responsible for majority of cases in all the three types of sexual assault reported. This is similar to reports from other places including India and USA by Jaya and Hindi (2007) 25  In response to a question item in our survey which state, "in your opinion why do men sexually assault women"? Among the reasons given by the respondents are to show domination, because men wanted to enjoy themselves, wickedness and punishment for women who exposed their bodies (badly dressed). Some said it was done to avenge insults that men received from some women when they proposed love relationship to them. These reasons were similar to reports from elsewhere. National Center for Victims of Crime. When respondents were asked for solutions to the problem of sexual assaults majority advocated enforcement of the laws on sexual offences, in addition they admonished Ladies to avoid exposing their body by their mode of dress and avoid abuse of alcohol and drugs. Conclusions: Sexual assault against women is a public health problem with serious negative effects on the health of women both in the urban and rural areas of Osun State. Women in the rural area suffered more repeat assault, while living in urban area and having male partner are associated with female sexual assault. Implication for intervention programme.. Population based (rural and urban) primary prevention education programme, highlighting negative cultural practices amongst others should be targeted at the youths, while emphasizing health hazards of sexual assaults against women. Intervention to encourage reporting of sexual assault by the victims and prosecution by the polices is desirable.

Conflict Of Interest.
The study was funded by TEt-fund 2013.

Appreciation:
The Authors appreciated the respondents who shared their guarded personal unpleasant sexual experiences with our research team.

Acknowledgement:
The logistic support from officers in Osun State Office of National Population Commission. Authors contribution: Adeleke, N.A. Conceived the idea of the study, design of the study, participated in the field work, analysis and manuscript writing. Adebinpe O.W. Participated data analysis and report writing. Farinloye E.O. Participated in field work, data collation, and manuscript writing. Olowookere A. S. Participated in study design, data analysis and manuscript writing. Decision To Publish: All authors approved the final manuscript and consented to publishing the study in Bangladesh journal of medical sciences.