Contraceptive Use among Gender Based Violence Survivors: The Case of Matabeleland South, Zimbabwe

Zimbabwe’s contraceptive utilisation has improved significantly, but this progress is not uniform across all its provinces. Matabeleland South in particular, is one province with the lowest contraceptive prevalence, against a background of significantly high spousal violence. This study sought to investigate modern contraceptive prevalence among GBV survivors in rural Matabeleland South. The study used cross-sectional data collected from 130 female survivors of Intimate Partner Violence (IPV) from Bulilima and Umzingwane districts in Matabeleland South. The study found the modern contraceptive prevalence to be 56%, with the most commonly reported contraceptive method being the oral pill, followed by injectables and implants, condoms, and lastly, female sterilisation. The study also found that education, age, and level of health care access had a positive association with contraceptive use, whilst age difference with a partner and marital status had a negative association with contraceptive use. Overall, the study indicates that contraceptive utilisation is still low amongst GBV survivors in Matabeleland South. Programming aimed at improving contraceptive use should therefore emphasize increasing education among young women to raise awareness on the benefits of using contraceptives, as well as involving spouses of survivors in programming to facilitate rapid adoption of contraceptives.


Introduction
Contraceptive use is a key reproductive health indicator due to its important role in improving reproductive health outcomes and consequently sustainable development. Where women and girls have adequate access to family planning services, they can control when and how often they have children; in turn, it improves women's access to education and income opportunities (Care International Australia, 2017;UNAIDS, 2019;UN WOMEN, 2016). Despite the importance of contraception in the development conversation, contraceptive use remains low, particularly in rural areas. WHO (2016) reported that the unmet need for contraception remained high, with 225 million women failing to access needed contraceptives and 74 million women having unwanted Contraceptive use is a key reproductive health indicator due to its important role in improving reproductive health outcomes and consequently sustainable development. Where women and girls have adequate access to family planning services, they can control when and how often they have children; in turn, it improves women's access to education and income opportunities (Care International Australia, 2017;UNAIDS, 2019;UN WOMEN, 2016). Despite the importance of contraception in the development conversation, contraceptive use remains low, particularly in rural areas. WHO (2016) reported that the unmet need for contraception remained high, with 225 million women failing to access needed contraceptives and 74 million women having unwanted pregnancies. UN (2017) and UNFPA (2020) also reported that globally, one in ten women married or in a union have an unmet need for family planning. In lowmiddle income countries, an estimated 232 million women are not using contraception despite avoiding pregnancy. In Sub-Saharan Africa, childbirth and pregnancy are the leading causes of death amongst adolescents (Ngome, 2016;Garrene, 2015). Furthermore, levels of access and utilisation of contraceptives are constrained by high rates of gender-based violence (GBV), which are prevalent in Africa (UNAIDS, 2019;UNDP, 2019). UNESCO (2013) defines GBV as: "Violence directed against a person because of his or her gender expectations of his or her role in society or culture" (UNESCO 2013: page 12) In Sub-Saharan Africa, GBV ranges from 27%-50%, with spousal violence having the highest prevalence (Bannister, 2014;UNFPA, 2018). Women in abusive relationships are often actively prohibited from using contraceptives (UNAIDS, 2019). This, in turn, exposes women at risk of unplanned pregnancies, which results in poor investment in the child's health, as well as poor mother and child health outcomes. Studies have revealed that the poor use of contraceptives is associated with limited choice in contraceptive methods, as well as accessibility issues, especially in rural or remote areas (Muanda et al., 2017;Li et al., 2019).
Zimbabwe is one of the countries that has been successful in meeting its target of a contraceptive prevalence of 68% by 2020 by the Family Planning (FP) 2020 initiative. Zimbabwe's contraceptive prevalence rate stands at 69%, making it one of the highest in Sub-Saharan Africa (UNFPA, 2018). However, some provinces, such as Matabeleland South, lag behind the national average in terms of contraceptive use. Matabeleland South has the lowest contraceptive prevalence rate in the country, which stands at 60%. Researchers such as Muanda et al. (2017), Ndayizigiye (2014), Li et al. (2019), Kebede et al. (2019) have noted that in-country variations exist in terms of contraceptive use, among many countries in Africa, with women living in rural areas having lower modern contraceptive prevalence rates than those residing in urban areas. This plight is further compounded by high rates of spousal violence, where the national average for spousal violence is 35% whilst that of Matabeleland South is 32.4% (ZIMSTAT, 2016). For a country like Zimbabwe, such a trend is concerning, given that more than 60% of its population resides in rural areas (ZIMSTAT, 2012).
Nonetheless, not many studies have been conducted to assess the contraceptive prevalence and reasons cited for not using contraceptives by women living in such areas. This study sought to assess the modern contraceptive prevalence rate, as well as the most prevalent forms of contraceptives used amongst GBV survivors in Matabeleland South, to facilitate a better understanding of the birth control methods available to GBV survivors in rural areas. This will help inform policies to improve reproductive health outcomes uniformly across the country.

Review of Literature
Contraceptive prevalence has been investigated by several authors, who found results consistent with their different contexts. Below is an overview of studies that assessed trends and predictors of contraceptive use. Tolefac (2018) conducted a study to assess trends and patterns of family planning methods among women in rural Cameroon. His findings revealed that contraceptive use stood at 17.4%. The study revealed that implants, intrauterine devices (IUD'S) and condoms were the most common contraceptive methods. Furthermore, age, being married and education were positively associated with the use of contraceptives. Similar results were also found by Palamuleni (2013), who found education age, marital status, spouse's approval of family planning method and number of living children to be positively associated with contraceptive methods. Kebede et al. (2019) conducted a study to assess demand for modern contraceptives amongst rural women in Ethiopia, using a cross-sectional study design and a multivariate logistic regression for analysis. The study results indicated that modern contraceptive prevalence was at 55.7%, with 64.3% using implants, 23.1% using injectables, 1.8% using pills, whilst the remaining used long-acting permanent methods. Age, education, husband not wanting more children, and several children were positively associated with contraception. Some of the reasons cited for not using contraceptives were, wanting to have more children, fear of side effects such as menstrual disturbances, and lack of adequate knowledge concerning how contraceptives work.
In a multi-country study conducted by Asaoula et al. (2020), the modern contraceptive prevalence was 22.7, 33.2, and 68.9% in Nigeria, Ghana, and Kenya, respectively. The study revealed that contraceptive prevalence was higher among women who had access to healthcare services either through nearby polyclinics or hospitals or health workers who paid them home visits. The results were consistent with those of Babazedah et al (2019) in DRC, who found that improving healthcare access through increasing the number of facilities stocked with contraceptives improved their use.
A study done in Ethiopia by Kitila et al. (2020) revealed that spousal age difference affects contraceptive use negatively. Women with spouses who were 10 years older were less likely to use contraceptives. These findings are also similar to those found by Ibisomi (2014) in Nigeria. Other factors that were significant in predicting contraceptive use in this study are the husband's education, place of residence, wealth status, employment and religion. Peer et al. (2013) assessed factors hindering contraceptive use in rural South Africa and found the mCPR 39.3%. Parity, male entitlement, and high self-esteem were positively associated with the use of contraceptives. Makhola (2019) also studied predictors of contraceptive use in South Africa and found education and having a sexual debut at 15 or older to be positively associated with contraceptives. On the other hand, being married, not having given birth before and living in rural areas were negatively associated with contraceptives. Other authors, such as Muanda et al. (2017) and Sedgh et al. (2016) found similar results. Additionally, Sedgh et al. (2016) found that opposition by the spouse, lactational amenorrhea and poor healthcare access were barriers to contraceptive use.

Materials and Methods Data Sources
This study used primary data collected from female survivors of intimate partner violence (IPV) or spousal violence, aged between 15 and 49 years. This was because they are the most vulnerable to poor utilisation of sexual and reproductive health services. The study adopted a cross-sectional design, and a survey was used to collect qualitative and quantitative data. Descriptive methods were used to ascertain the level of utilisation, the prevalent methods of contraception, and reasons for not using contraceptives amongst the GBV survivors. The study also used the data collected to assess the determinants of contraceptive use through logistic regression.

Sampling
The sample size for this study was determined using the Cochran formula (1977) formula for sample determination: n = z 2 pq/e 2 Where n is the sample size, z is the selected critical value of desired confidence level of 99% in this study, is the proportion of an attribute that is present in the population (proportion of women in an intimate relationship who are experiencing violence). According to ZIMSTATS (2016) Demographic Health Survey, the proportion of women in an intimate relationship experiencing spousal violence in Matabeleland stood at 32.4%. Therefore p is 0.324 and q=1-p=0.676 is the estimated proportion of the population that has experienced IPV and eis the allowable error, which is equal to ±10% in the present study, while z value is 2.57. Therefore the sample size for the study was calculated to be 144 female GBV survivors.

Sampling Procedure
This study used a sample purposefully sampled from Umzingwane and Bulilima to represent the province as the districts have a high number of organisations actively working on GBV issues and women's rights. Purposive sampling was a useful technique whose application was proper since the target population was survivors of GBV, which is a very sensitive matter. Obtaining responses from this delicate population, therefore, required engaging organisations whom the survivors trusted and were willing to open up. As such, Umzingwane and Bulilima were ideal for this study. Ward selection was made randomly, and 3 wards were selected for both the Umzingwane and Bulilima districts.
Questionnaires were administered to all the women present who fit the criteria of being aged 15-49, and were in a union or relationship. GBV survivors were identified by whether or not they had indicated experience of any form of GBV from their spouse on the GBV section of the questionnaire. Active users of modern contraceptives indicated using any modern contraceptive for at least a year.

Ethical Considerations
The research followed ethics that conformed to the ethical guidelines prescribed by the Department of Economics of the University of Zimbabwe. The researcher obtained a formal letter from the Department of Economics granting her permission to collect data in the Matabeleland South Province. Furthermore, the researcher availed all information regarding the research study's objectives to the respondents. Before completion of the questionnaires, the researcher gave an account of who she was, the purpose of the research and the objectives of the study. The researcher sought consent from the respondents and explained clearly that the findings would be used purely for academic purposes and that confidentiality would be adhered to in handling the information provided.

Data Management and Analysis
After collecting the data, the researcher screened the responses collected from all the administered questionnaires and selected questionnaires where a woman had indicated that she experienced spousal violence for analysis. Questionnaires were checked for completeness and consistency before data were then entered into statistical software STATA for analysis. The response rate for the survey was 90.3% and the total number of respondents was 130 survivors. Data were analysed using descriptive statistics and logistic regression. Descriptive statistics showed the background characteristics of the respondents, the most prevalent forms of contraception and reasons cited for non-use of contraceptives. Logistic regression was run to analyse the determinants of contraceptive use among GBV survivors, using odds ratios.

Results
The section presents results from the data analysis for the study, a description of the background characteristics of the respondents, then an analysis of the prevalence of modern contraceptive use, the type of modern contraceptives and reasons for not using contraception cited by those who highlighted they are not using contraceptives. A logistics regression was also run to establish determinants of contraceptive use among the respondents. The response rate for the survey was 90.3%, and a total of 130 respondents completed the questionnaire. Table 1 shows descriptive statistics for the continuous variables used in the study. The average number of years spent in school was 10 years whilst the maximum was 17years. This shows that the majority of respondents had acquired education up to secondary level only. The average number of children for the group of respondents was 4, with the maximum being 7. This reflects the high fertility rate in the area, given that the average number of children per woman was 4. The respondents' average age was 20 years, whilst the maximum was 49 years. The spousal age difference was at a minimum of 1 year, a maximum of 30 years, and an average of 6 years. The age difference with partners shows intergenerational relationships in the area, as evidenced by the high spousal age differences. Source: Author's Computations N=130 Table 2 below shows the descriptive statistics for categorical variables used in the study. Religion had 5 categories, and the Apostolic sect had the highest number of respondents at 42.3%, followed by the Pentecostal which had 26.2%, then the None/another category with 12.3%, the Roman Catholic with 10.8% and lastly the Traditional sect with 8.5%. The findings also show that 88.5% of the respondents were married, whilst 11.5% were single but intimate. The data shows that 57.7% of the respondents had high access to media, whilst 42.3% indicated otherwise. 51.5% of the respondents indicated having high access to healthcare services. The data also shows that 60.8 % of the respondents had high levels of autonomy.

Contraceptive Prevalence
Contraceptive use reduces the risk of exposure to reproductive health issues such as unwanted pregnancies, sexually transmitted infections, including HIV and abortions. Findings from the study revealed that 44% of the GBV survivors reported using contraceptives, Figure1.

Types of Contraceptives
Respondents who reported that they had used contraceptives were asked to state the contraceptives they had used. The majority of the respondents, 43% reported that they had used the pill (Table  3). The implant and injectable were each reported by 21% of the respondents. A small proportion of the respondents, 11%, reported that they had used a condom. Only a paltry 3%, of the respondents, reported using emergency contraception.

Source: Author's Computation N=73
Reasons for Not Using Contraceptives GBV survivors who had reported that they had not used contraception were asked the underlying reasons. About 43% of the respondents reported not using contraception because they were breastfeeding (Figure 2). A significant proportion of the respondents, 29%, reported that they were afraid of the side effects associated with contraceptives. About 15% of the respondents reported that they had not used contraceptives because of the disapproval by the spouse. A small proportion of the respondents, 7%, reported a lack of knowledge.

Source: Author's Computations N=57 Figure 2: Percentage Distribution of the Reasons Why GBV Survivors Did Not Use Contraceptives
The table below shows the results of logistic regression of contraceptive use in the form of odds ratios. The results revealed that education, age and healthcare access had a positive association with contraceptive use. Marital status and age difference with the partner on the other hand had negative associations with contraceptive use.

Determinants of Contraceptive Use
This study found positive associations between healthcare access and the use of contraceptives. This resonates with the findings such as Stephenson et al. (2008) in South Africa, Babazadeh et.al. (2020) in Kinshasa and Wang et.al (2012) in a multi-country study done in Tanzania, Kenya and Rwanda. Asaoula (2019) found that women with better access to healthcare services were 1.9 times more likely to use contraceptives. This compares with the findings of this study, where better access to healthcare improved the odds of using contraceptives by 71%. Where contraceptives are easily accessible, especially for survivors of GBV, it gives them an incentive to use contraceptives as they also have readily accessible information about the merits and demerits of each method. This means that interventions to improve the utilisation of contraceptives should improve the accessibility component. Increasing funding allocated to providing sexual and reproductive health services, especially contraceptives, would allow contraceptives to be easily accessed by female GBV survivors. Interventions such as increasing the presence of mobile clinics, or centres where survivors can access reproductive services such as contraception, would facilitate greater uptake of contraceptives amongst survivors.
Marital status was observed to have a negative association with contraceptive use amongst survivors. The odds of married survivors using contraceptives was 24% lower than survivors in a single but intimate relationship. Similar results were obtained by researchers such as Yadav et al. (2017), in India, Subedi et al. (2018 in Nepal and Makola et al. (2019) in South Africa. In a marriage institution, norms that exist in rural areas peddle male superiority and dominance. As a result, survivors have limited room to contest their spouse's decision should they decide that he does not want them to use contraceptives. Using them covertly would also be regarded as challenging the husband's authority, as such survivors who are married are less likely to use contraceptives compared to those in intimate but single relationships. To improve contraceptives among survivors, it is important that development partners and policymakers engage in campaigns to end child marriages. In this way, women have a late sexual debut, and commit to marriage at a point where they have a better appreciation of the demands of marriage, as well as the better capacity to select partners who will not compromise their reproductive health.
Age difference with a partner was found to have a negative association with the use of contraceptives amongst female GBV survivors. Similar findings were obtained by researchers such as Ibisomi (2014) in Nigeria and Kitila (2020) in Ethiopia. In this study, a one year increase in age difference with partner decreased the odds of using contraceptives by 16 %. Amongst rural populations, huge age differences between spouses are common, more so in rural areas where early child marriages are a common phenomenon. The age difference between spouses mean there is gap in maturity and experiences, which usually introduces conflict on issues such as reproductive health. Where a survivor's husband is older, she runs the risk of abuse of her rights to contraceptives as she is not in a place to fairly negotiate with her much older spouse. From a programming perspective, there is a need to educate young women on the dangers of intergenerational relationships and where possible, encourage them to avoid getting married too early. This will allow them to first mature to a point where they can make sound decisions regarding their reproductive health and marriage. Such information goes a long way in ensuring that young women are fully equipped to make decisions that are not from a position of vulnerability.
Education had positive associations with contraceptive use among GBV Survivors. Survivors who were educated had 43% higher odds of using contraceptives. Palamuleni (2013), Babazedah et.al (2020, and Rutaremwa and Kanagenyi (2017) got similar findings in Malawi, DRC and Uganda, respectively. Education has the effect of improving one's decision making, including decisions concerning one's health. As such, survivors who are more educated are better positioned to make reproductive decisions than those with fewer years of education. They are also able to understand better the implications of not using contraceptives and the benefits should they do so. This means policies that encourage higher education should be encouraged as this creates an inherent capacity in the female survivors to process information with a realisation of the consequences of not using contraceptives and any other sexual and reproductive health services, especially amidst violence.
The age of the respondents showed a positive association with contraceptive use. The same observations were made by researchers such as Kebede et al. (2019) in Ethiopia, Osmani et al (2015) and Terrani et al., (2001) in Tehran. This study revealed that a one year increase in age was associated with a 36% increase in using contraceptives, amongst GBV survivors. This observation is attributed to the fact that most women are more likely to want to give birth in the early years of their life, while at later stages, they are likely to have reached the desired family size. Moreover, the pressure of proving fertility is high in the early ages, hence the increased use of contraceptives in later years. The importance of age in determining contraceptive use in the current study means that interventions aimed at promoting the use of contraceptives should target the young female survivors of GBV to reduce chances of unwanted pregnancies and also exposure to sexually transmitted infections, including HIV. Increasing the presence of youth-friendly services in remote areas will also facilitate increased uptake of contraceptive use among younger women. This reduces the risk of teenage pregnancies and maternal deaths associated with poor contraceptive use among young women.

Contraceptive Prevalence and Types of Contraceptives Used
Modern contraceptive prevalence (mCPR) in this study was found to be 56%, a figure which is much lower than the national mCPR, which stands at 69%. Similar results were also obtained by Lasong et al. (2019) and Peer et al. (2013), who found mCPR to be 43% in rural Zambia, and in rural Western Cape South Africa, where it stood at 39.3% in comparison to the then national average of 55% and 64%, respectively. Rutaremwa and Kabagenyi (2016) also conducted a study among rural women in Uganda where they found the mCPR to be 30% against a national average of 36%. Other studies such as Li et al. (2019) also found subnational variations in mCPR across Burkina Faso, Kenya, Uganda, Ghana and Ethiopia. As such, this observed disparity points to inequity in terms of contraceptive access. To curb this challenge, policymakers should prioritise an increase in investment in contraceptive access in rural areas such as Matabeleland South. As of 2015, rural clinics and hospitals only received 5% of Zimbabwe's total health budget, yet more than 65% of Zimbabwe population resides in rural areas (UNFPA, 2018;ZIMSTATS, 2012). Findings This study therefore reinforces the need for policy to address such gaps in accessing health care.
The most common modern birth control method (MBCM), from this study, was the oral pill, followed by implants and injectables, condoms, and lastly, emergency contraception and female sterilisation. In Cameroon, Tolefac et al. (2018) found that implants and intra-uterine devices were the most common contraceptives among rural women, whilst in rural Ethiopia, Tsehaye et al. (2013) found injectable and oral pills to be the two most preferred methods for contraception, followed by implants and lastly female sterilisation as in this current study. Lasong et al. (2019) also conducted a study assessing contraceptive use in rural Zambia. They found the injectable having the highest prevalence followed by implants, whilst female sterilisation was the least used method as in this current study. Contrary to most studies that show injectable, a long-acting contraceptive as a dominant contraceptive method, this study finds that the pill or oral contraceptive is the most common birth control method among GBV survivors in Matabeleland South. This could be because accessing long term acting methods is hindered by accessibility challenges, given that only 52% of the respondents cited high access to health care facilities. It is therefore imperative for government to make contraceptives more accessible, especially to the demographic of GBV survivors as their access to such services might be constrained. Programming to improve contraceptive use should target an increase in long-acting reversible contraceptive methods as these have more efficacy than short term methods.

Reasons for not Using Contraceptives
This study also revealed the reasons why survivors did not use contraceptives. The most dominant reason was breastfeeding, followed by side effects, opposition by partner, lack of knowledge and other reasons, respectively. These findings were supported by several researchers such as Sedgh et al. (2016) noted that breastfeeding was cited as a cause for not using contraception in African countries. His findings revealed that the percentage of women who shared this reason as a cause for not using contraception ranged from 40% in Mali, 38% in Burundi, Niger and Nigeria, 25% in Uganda and 22% in Zimbabwe. This finding is consistent with the high fertility rate observed in Matabeleland South. Due to cultural norms and expectations of having many children, which is observed in many rural communities, there is a high probability that most women will be breastfeeding as they nurse their many children. As they breastfeed consistently, the women stop ovulating and consequently use lactational amenorrhea as a method of contraception.
Fear of side effects and lack of knowledge were also cited as reasons for not using contraception. This is consistent with Kebede et al. (2017) in Ethiopia and Muanda et al. (2017) in DRC. Kebede et al. (2019), Muanda et al. (2017), also noted that women who did not use contraception cited side effects, lack of knowledge, and opposition to use as reasons for not adopting modern contraceptives methods in DRC, and Ethiopia respectively. This points to a need to ensure proper dissemination of information to women in rural areas on how contraceptives work to facilitate their rapid adoption. The government should therefore place focus on incorporating knowledge dissemination to remote areas so that decisions regarding contraception are made with full information of how they work and the benefits associated. In this way, the decision to use contraceptives is not based on myths or incomplete information.
The opposition of the use of contraceptives by partners, which was cited in this study as a reason for not using modern contraceptives is consistent with other studies such as Ndayizigiye et al. (2014) in Burundi, Kriel et al. (2019) in South Africa, and Sedgh et al. (2016) in a multi-country study. Women indicated that opposition by partners or lack of male involvement discouraged their use of contraceptives. In rural communities such as Matabeleland South, patriarchy is dominant. As such, continued use of contraceptives without the approval of the male would imply that the woman is challenging the authority of their partner and make her susceptible to even more violence. In light of this information, it is imperative for programming to encourage contraceptives to involve males to facilitate smooth adoption of modern contraceptives by women, with the support of their partners. With a better understanding of the benefits of contraceptive use to families, the male spouses of GBV survivors are more likely to support their partner's choice to use contraceptives.

Conclusion
This study revealed that contraceptive use among GBV survivors was much lower than the national contraceptive prevalence rate. The study also noted that the pill was the most preferred contraceptive method amongst female GBV survivors in Matabeleland South. Side effects, breastfeeding, opposition by partner, lack of knowledge were cited by female GBV survivors not utilising contraceptives. Age, marital status, age difference with partner, education and healthcare access were also identified as determinants of contraceptive use among GBV survivors. This study revealed that age, education, and healthcare access had positive associations with contraceptive utilisation whilst marital status and age difference with a partner had a negative association with contraceptive use. Policymaking to improve reproductive health care should therefore take into cognisance the challenges associated with contraceptive use, which are peculiar to survivors in crafting programs aimed at increasing contraceptive use.