Memoire Online - Factors associated with pregnancy termination in Cote d'Ivoire using the demographic health survey-multiple indicator cluster survey data for Cote d'Ivoire from 2012

Title: Factors associated with pregnancy termination in Cote d'Ivoire using the Demographic Health Survey-Multiple Indicator Cluster Survey data for Cote d'Ivoire from 2012

Pennsylvania State University (Penn State), College of Medicine, Hershey, Pennsylvania, USA

Pregnancy termination is used as family planning by women in many societies and countries of Africa,but is often conducted in unhygienic and unsafe circumstances. Abortion rates are increasing every year making this practice a significant contributor to maternalmorbidity and mortalityin sub-Saharan Africa, including Cote d'Ivoire.

This study objective is to explore the factors associated with pregnancy termination.

This study used the data from the Demographic and Health Survey (DHS) conducted in 2012 in Cote d'Ivoire. Univariate and multivariate logistic regression models were used to identify factors associated with a life time history of having had one or more pregnancy terminations.

A total of 10,848 women aged of15-49 participated in the survey. The majority of the participants were married (64.2%) and nearly three fourths had born at least one child. The rate of current contraceptive use was low (18.8%) and 18.03% reported having had one or more pregnancy terminations. Women who were married, Christian, educated and wealthy were more likely to have had a reported pregnancy termination. In addition, a woman who was older and had born at least one child was more likely to have had a termination. Women who had a history of pregnancy termination were more likely to report that they currently used contraception.

The results of this study indicate that women of childbearing age in Cote d'Ivoire have limited access to contraception and some have sought pregnancy termination. Women of higher meansin terms of wealth, education and working status are more likely to have had a reported pregnancy termination.

Keywords: pregnancy termination, abortion, Africa, maternal mortality, Demographic Health Survey.

Abortion is a major contributor to maternal mortality in sub-Saharan Africa. Over 6 millionabortions were reported as induced abortions inAfrica with more than 1.8 million occurring in West Africaduring 2008 (Guttmacher, 2015). In 2008, the maternal mortality ratio in West Africa was 28 deaths per 1,000 live births and the rates are still increasing every year (Guttmacher, 2015). The maternal mortality ratio is higher in Cote d'Ivoire than West Africa as a whole with 400 maternal deaths per 100,000 live births (DHS, 2012).

Despite abortion beingillegal in almost all African countries,9% of maternal deathswere attributed to abortion in 2014 (Guttmacher, 2015).For this purpose,in Africa, pregnancy termination (induced abortion, spontaneous abortion or miscarriages) specifically represents a serious public health problem because they are often conducted in unhygienic and unsafe circumstances (Varga, 2002). It is estimated that 14% of maternal deaths are attributed to unsafe abortion and to pregnancy complications (WHO, 2011).According to AIDS Indicators Survey of Cote d'Ivoire (AISCI-2005), pregnancy termination such as abortion is the second cause of maternal mortality and the rates of abortion are still increasing every year (WHO, 2010). Thus, this situation increases the burden of maternal mortality and morbidity. It appears that stigmatization, shame and societal pressure drive practices related to abortion in many Africans societies(Adjei, 2015).

Despite the promotion of family planning programs, women continue to practice abortion in many African countries (Guttmacher, 2015). Because of its illegality, the majority of terminated pregnanciesare clandestine and many are unsafe(Varga, 2004). A study carried out in Kenya showed that the main reasons of abortion reported by women are divorce, poverty, loss of job and social exclusion, loss of marital viability(Izugbara, 2009).

In Cote d'Ivoire, researchers have addressed issues surrounding induced abortion, but many related to factors associated with termination of pregnancy remain insufficiently explored(Svanemyr&Sundby, 2007).Among the available studies, few have identified the factors associated with pregnancy termination (Goyaux, et al., 2000). Although the study achieved by Goyaux et al. (2000), on abortion in Abidjan (Cote d'Ivoire) showed that abortion is fairly well documented in almost African countries, it is less well documented in Cote d'Ivoire. Thus, it is crucial to investigate the factors that can be associated to pregnancy termination in this country. According to UNICEF (2012) the contraceptive prevalence was 18.2% in 2012 in Cote d'Ivoire, drives to high rate of abortions.

Given the foregoing, thisstudy on pregnancy termination in Cote d'Ivoire, aims to analyze the factors involved in the practice of pregnancy termination.

The mainobjective of this study is determine the chain of factors associated with pregnancy termination in Cote d'Ivoire.Specifically, it is to:

· Identify the profile of women of childbearing age (15-49) who have had pregnancy termination);

The purpose of this research is to contribute to improving knowledge on the determinants of pregnancy termination, which will help the government to implement significant reproductive health programsand plan new strategies to ameliorate thepopulation health.

H1: Socio-cultural (age of the woman, marital status and religion) factors explain the recourse to pregnancy termination in Cote d'Ivoire.

H2: Socioeconomic factors (occupation of women and education level and household wealth) explain the recourse to pregnancy termination in Cote d'Ivoire.

1. The study is located in Cote d'Ivoire a West Africa country.All women aged 15-49 years usually living in the selected households, or present the night before the survey were eligible to be interviewed.Specifically, the study is about women of childbearing age, 15 to 49, who reported that they have had a pregnancy termination. Data examined in this study were obtained from Demographic Health Survey of Cote d'Ivoire database for the year 2012.

The combination of spontaneous, induced abortion and miscarriage into one variable (pregnancy termination) does not give specific information on abortion for example. With such a variable, it is difficult to focus on a specific topic such as induced abortion. Also, there are problems of estimations because, sometimes, participants don't know really know the difference between induced, spontaneous abortions and miscarriages. Moreover, it can be difficult for all the respondents to accurately report some of the past pregnancy outcomes.One of the limits of the Demographic Health Survey (DHS) is that there is a period of 5 years between two surveys.

Pregnancy termination. A termination of pregnancy is an induced or spontaneous abortion that occurs after 20 weeks of gestation (The Johns Hopkins Manual of Gynecology and Obstetrics, 2011). The termination of pregnancy is formed of three items which are induced abortion, spontaneous abortion and miscarriage.

Miscarriage. Miscarriage may be definedas involuntary termination of pregnancy before the fetus is viable. (The Johns Hopkins Manual of Gynecology and Obstetrics, 2012)

Induced abortion.Induced abortion is the termination of pregnancy or expulsion of the fetus before the fetus is viable during 28 weeks. In fact, an induced abortion occurs when a woman makes a choice to stop a pregnancy process. (The Johns Hopkins Manual of Gynecology and Obstetrics, 2011).

Spontaneous abortion. Spontaneous abortion, also known as miscarriage, is the unintentional or involuntary expulsion of an embryo or fetus before the 24th week of gestation (Borgatta, 2014).

Socio-cultural factors. These factors represent customs, attitudes, beliefs, values and perceptions that determine the behavior of individuals in society. Every society has its norms and values ??regarding sexuality. Thus, the mean of cultural factors is a combination of factors such as norms and values having a direct or indirect effect on pregnancy termination. (Rwenge, 2002).

Sociodemographic factors. This is essentially the woman's age, marital status and number of children of a woman. Indeed, we womanmust reaches a certain age for childbearing expect to be exposed to the risk of pregnancy, as part of this study will be of age 15 years. According to the literature, marital status influences the practice of abortion in several ways.Single women are more likely to use more to abortion than married women. Also, some studies have found a relationship between age and number of children of a woman. Abortion is used by the younger generations todelay childbirth, while olderwomen use it to space births (Adjei, 2015).

Socioeconomic factors. These factors represent the ability of a woman to get the financial means to use modern methods to prevent pregnancy. It is, also, the ability of a woman to take care of a child and to manage her economic and financial status. The indicators of these factors are the occupation of women, the household wealth, the employment status (occupation) and education levels (Moyabi, 2010).

Contraceptive practice. Contraception practice represents the use of a modern contraceptive methods to avoid unintended pregnancy (Rwenge, 2002).

Lack of statistics and data on pregnancy outcomes in Africa limitthe literature review. According to some surveys and studies, the recourse to pregnancy termination in Africa is associated withfactors such as demographic, socio-cultural, socio-economic and institutional. To better understand the practice related to pregnancy termination, the review of literature is divided under the following subheading. First of all,it will be determined the differentfactorsapproaches such as sociocultural, socioeconomic, institutional and health and then in second hand, it be will identified the different determinants associated with the pregnancy termination (variables).

Adjei, (2015) showed that women practice abortion at different ages of their life. Many women who have more than 5 children are likely to limit their births (cite). In Gabon, the variation of the prevalence of pregnancy termination whichis 15% and 23% in women over 20 years is lower among younger women (6%)(Mendame, 2005). When analyzing the age of women at their first abortions, it appears that 44% occurred before the age of 20 years (Barrere, 2001). In Yaoundé and Douala (Cameroon), a survey of 1,638 women revealed that women aged of 24-35 years have more miscarriages (30.3%), followed by those aged 45-49 years (29.7%). Women aged of 25 years reported having had an abortion (12%) in both localities (Ngwe et al. 2004). In Ghana, the average age ofabortion is about 30 years, but it seems that this practice is common for young single women (Varga, 2002). According to Adjei et al. (2015) young women are more likely to practice abortion than older women in Ghana. In Bamako (Mali), a survey conducted about abortion practice in health centers showedthat, it is especially women under 25 years old and single having recourse to pregnancy termination (Konate, 1993). Abortion is often practiced by young women who, by this practice, interrupt their first pregnancy through an abortion(Mote et al., 2010). A survey conducted in Gabon highlighted the strong prevalence of abortion among adolescents and young people (Bibang, 2010). Also in Egypt, women hospitalized for illegal abortions are young women, unmarried, uneducated, educated and without children. (Ezzeldinet al., 2010).

It is in the same way, Mendame (2005) found that the distribution of women by abortion and age show that the proportion of women who have already practiced abortion is highest among 35-49 years (75.1%), followed by those of 25-34 years (59.2%) and then by those of 15- 24 years (53%).Also, in Accra in Ghana, according to a study in a hospital, a quarter(25%) of women less than 20 years from birth to the second pregnancy had interrupted their first pregnancy unsafely(Mote et al., 2010).We see that all women (15-49 years) practice pregnancy termination but literature reveal that there is a cultural cause of abortion which is the fear of social or family rejection.

In many African societies,sexuality represents an aspect of social,cultural behavior and is qualified as a taboo. Thus, norms in terms of sexuality are in many social environments are strict and stigmatized(Izugbara, 2009). In a perspective of religious morality, sexual intercourse should occur for example in the context of a socially recognized union and should be reserved exclusively for procreation (Rwenge, 2002).Adjei et al. (1999) revealed that premarital pregnancy is stigmatized and severely punished.The pregnancy of the girl is stigmatized in manyAfrican societies. Societal pressure, shame, family rejection represents the causes of abortion practice (Izugbara, 2009). It leads, very often, to illegal and unsafe abortions. The fear of the reaction of parents or family facing a pregnancy considered unacceptable (young women, single or deemed not to have sex) or issues of partners (disagreements, premarital pregnancy, refusal of paternity, occasional partners, adulterous pregnancy) can explain some pregnancy terminations (Maina, 2015). These patterns depend on the age of women and their marital status. In fact, the decision to abort for young women largely dependent on the reaction of the partner and his ability to accept the paternity of a pregnancy (Svanemyr, 2007).

Studies in Cote d'Ivoire emphasize the importance of familial reasons in decision to practice a pregnancy termination such as induced abortions. The main reasons found are unstable unions, the fact of refusing to recognize a pregnancy by a partner, marriage problems, and fear of scandal or reaction of parents (Izugbara, 2009). In Douala and Yaoundé, "the fear of being rejected by parents or family" was cited as the primary cause of abortion among young girls (52.8%) (Ngwe et al. 2005).In Bamako(Mali,) a country dominated by Islamic religion, the main reasons of abortions reported by women are religious (38 to 64%), family pressure (31 to 45%) and the refusal of their partner to recognize a pregnancy (1.7 to 14.6%) (Konaté et al. 1999).The fear of parents is cited as a main reason for abortion, especially by young women: 16.4% of women under 20 years old had abortions for this reason in Gabon 22% in Benin 71% in Togo 17% in 1996 to 26% in Nigeria and Uganda (Adjei, 2015 and Mendame, 2005). In Ghana, 21% of women mentioned fear of parents and 20% fear criticism and severe punishment of thefamily and neighborhood (Mote et al. 2010).

In the same way, the determinants and behavior that young couple must face to an unwanted pregnancy is a reason of high rate of abortions. The fact that abortion is illegal and the lack of money, many young people cannot count to their parents economically. Indeed, they can't receive assistance from their parents to take care of the child and thus parents push the young couple to practice abortions(Svanemyr, 2007).

Pregnancy termination specifically, the abortion is often practiced among young educated women and almost to those enrolled in secondary school who want to complete their education (Moyabi, 2002). In sub-Saharan Africa, it is reported that in many countries the majority of women who have abortions are teenagers want to continue their studies or work to improve their economic status, so that abortion is used to avoid or delay a pregnancy.(Izugbara, 2009).

Education is one of the main reasons by women when they practice a termination of pregnancy. In Cote d'Ivoire, for example, this reason was mentioned by nearly 55% of women across the country, 19% of them in Abidjan (DHS, 2012). The majority are educated of secondary school or higher and 36% of undergraduates (Guillaume, 2000). In Bamako, in Mali, "the school constraints" are among the top three reasons for abortion (Konaté et al. 1999). In Togo, education is mentioned by a third of women under 20 years as abortion pattern. (Djoke, 2004). This reason is also cited by almost of women aged 15 to 24 in Kenya (Izugbara, 2009). Also 26% to 38% of women have cited the same reason in Kenya (Maina, 2015).

Studies show socioeconomics factors influence the practice of unsafe abortion. (Mafuta&Beya, 2006). Indeed, economic difficulties conduct some women to practice pregnancy termination. These women highlighted the difficulties to take care of a child, incompatibility of women to simultaneously manage their occupation and the care to give to a child. (Mendame, 2005).

Also, economic difficulties and poverty are frequently cited among the reasons for abortion in most countries, such as in Nigeria (Izugbara, 2009). In Gabon, a third of women said to be facing economic difficulties or want to pursue professional activity (Bibang, 2010).Thus, in Douala and Yaoundé, the second reason cited by women to justify abortion is the lack of financial resources (39.6%) (Ngwe et al. 2005). In Uganda, women practice abortion in order to save their jobs or if their partner does not want to recognize the pregnancy (Kasolo, 2000).Unwanted pregnancies and abortion are the consequences in Africa of sex for profit or material. Indeed, some rich men who promise money, gifts to young women in exchange for unprotected sex, abandon them when a pregnancy occurs. Unable to meet the needs of the newborn, parents fear or pressure from these men, they are forced to abort these pregnancies.(Izugabara, 2009).

In summary socioeconomic factors influence the pregnancy termination in general. But, also, institutional approach has an influence on the practice of this event.

A restrictive legal framework on abortion. Abortion is still under restrictive measures including the majority of African countries.Today, abortion is totally banned in 15 countries (out of 190). In other countries, the situation varies from very strict limits to a very liberal policy in 55 countries (44% of the population), abortion on request or for economic and social reasons (Mendame, 2005).

In Francophone Africa (French speaking countries), the abortion laws are often still reminiscent of the French law of 1920, which condemn abortion and repress anyone practicing it. In English-speaking countries, the law is inspired by the British 1861 Act on offenses against the person. Thus, doing a pregnancy termination such as induced abortion is a crime and punished by the law (Bibang, 2010). But since 1999, in 26 countries, the termination of a pregnancy may be authorized only for the survival of the mother. (Adjei, 2015). In countries where abortion is liberalized, some restrictions still limit in practice. In Egypt, for example, health centers where these acts may be legally practiced are few, especially in rural areas (Ezzeldin, 2010). In South Africa, the law provides a legal age of pregnancy for an abortion (Guttmacher Institute, 2008). In Sudan and Zambia, the authorization of 2 or 3 doctors need to abort (Mote et al., 2002). All these restrictions penalize women. Instead of preventing abortions, they contribute to clandestine and dangerous practices. Thus, the high and increasing number of unsafe pregnancy termination isobserved in health centers of abortions complications. In West Africa, one million per year the number of illegal abortions are practiced outside any formal medical supervision. But the most affected continent region is Eastern Africa with 1, 34 million illegal abortions a year. (Guttmacher Institute, 2015).In South Africa, the law provides a legal age of pregnancy for an abortion (Mendame, 2005). In Egypt, for example, the authorization of 2 or 3 doctors is requiredto abort (Ezzeldin, 2010). All these restrictions penalize women. Instead of preventing abortions, women arestill practicingclandestine anddangerous pregnancy termination to avoid unintended pregnancies. It is in the same way that 68,000 women die of unsafe abortions every year.Thus, maternal mortality will occur every time those women face financial burden and social stigma of unintended pregnancy. But one of the factors of pregnancy termination is the lack of contraceptive methods.

Lack of contraception or failure of a method factor. Since the independences of many African countries (1960), severalprograms of reproductive health have been implemented through project for birth control. Abortion has been used as a birth control method before the use of contraceptive methods. The development of family planning programs has contributed to increase the use of contraceptive methods, butrates of abortions remain high.

In countries where contraceptive use is high, contraceptive failures are very common, by the persistence of unintended pregnancies which conclude with an abortion (Adjei, 2015). In this case, the resort to abortion may occur after a contraceptive failure, consecutive to misuse of a method or ineffective.In Africa, the lack of contraceptive methods drivesto contraceptive methodstounsafe abortion. (Izugabara, 2009).

Ezzeldin et al. (2010) conducted a situational analysis on abortion in Egypt. The study has showed that the main determinants of abortion are lack of avaibility of family planning and contraception services and lack of contraceptive methods knowledge. But the main determinant of unsafe termination pregnancy is the lack of contraceptive method.

In Cote d'Ivoire, illegal abortions persist for several reasons,mainly because of the lack contraception. (Moyabi, 2002). Thus, the use of contraceptive method is directly linked to pregnancy termination in case of induced and spontaneous abortion. (WHO, 2011).

However, in a conservative environment, medical providers do not accept to give the necessary advice to adolescent because of values and cultures. The consequence of such behaviors is the resultants of high rate of unintended pregnancies. They have proposed some solutions to reduce the high rate of terminated pregnancies. It is very necessary that contraception be accessible for active sexual teenagers. (Menachem et al. 2015).The post-abortion contraception can help reduce the need for abortion, but in many African countries it is difficult to get quality post-abortion care and requiredcounseling. Thus, the absence of reproductive health service and the inadequate use of contraceptive method represent a main cause associated to maternal morbidity and mortality. Also, health approach influences the pregnancy termination.

In many African countries, abortion is permitted if the physical or mental health of the woman is threatened. This pattern is not often cited by respondents in surveys and studies already conducted. In Gabon, 6% of women reported having used at least once to an abortion for health reasons (DHSG, 2000). This reason was cited by of women in Kenya (Izugbara, 2009), 3.2% in Douala and Yaoundé (Ngwe et al. 2005).

The different approaches reviewed in the literature allow us to identify the factors that might explain the recourse to abortion in Gabon.After determining the different approaches, it is important to identify the different variables associated with the pregnancy termination.

A number of variables related to socio-cultural, socio-economic and institutional influence the practice of abortion. These variables are sociocultural variables, religion, ethnicity, geographical setting, age, marital status, and number of living children.

Sociocultural variables. The practice of abortion is determined by specific standards and values ??of each society that influence the individual decision to practice abortion or not. Thus, sociocultural variables have an influence onpregnancy termination.

Religion. According Akoto(1985, cited by Djoke, 2004) «religion conveys a number of values ??and norms that govern the lives of the faithful on the behavioral and psychological level.» The majority of religions forbid abortion considered as a serious sin. This is the case of the Christian, Jewish and Muslim.However, these religions can allow an abortion in order to save the life of the mother (Rwenge, 2002).

Ethnicity. Many surveys have shown the effect of ethnicity on the pregnancy termination, but, numerous studies have documented differences. In Cameroon, for example, some sources have described the Beti-Fang as traditionally more tolerant vis-à-vis of sexuality norms and premarital childbearing than other ethnic groups (Calvès, 2004). But in Cote d'Ivoire, for example, women from the ethnicity Krou, are more likely to practice pregnancy termination such as abortion than the other ethnic groups (DHS, 1998).

Geographical setting.The place of residence is an important indicator of differentiation to explain the sexual behavior of women as they use or not pregnancy prevention methods. Several studies have shown that urban women were more sexually precocious than those living in rural areas. This can be explained as an aspiration of women to enjoy more freely of their bodies in the city than in the village (Bibang, 2010).

Age. Several studies have shown the correlation between the age of a woman and the practice of abortion. The majority of them related the high abortion rates among young singlewomenless than 25 years and particularly high before the age of 20 years (Adjei, 2015). But some researches showed that, women resort to abortion at all reproductive age (Maina, 2010).

Marital status.The marital status of women is an important determinant of abortion. Indeed, according to the literature more single women are more likely to terminate a pregnancy than single women (Izugbara, 2009). This is explained by the fact that premaritalpregnancy is assimilated to shame and social rejection (Adjei, 2015). Married women aborted either because they want children or to space their births (Izugbara, 2009). This practice can also intervene in the past as a result of an extramarital pregnancy. In Douala and Yaoundé, married women reported having an abortion in their majority by the lack of contraceptive methods (55%)and for economic reasons, including lack of financial resources (33.3%) and professional constraints (4.5%) (Ngwe, 2005). Divorce and loss of maritalviability represent a main cause of abortion, because these women do not have financial means to take care of an additional child. (Izugbara,2009).

Number of living children. The numberof livingchildrenof women also influences the practice of abortion. Some surveys carried out in hospitals suggest that a majority of women admitted due to post-abortion complications are young and are in their first pregnancy. However, other studies have found a greater diversity in the age and gender of patients. In a research conducted in a Kenyan hospital, for example, the majority of patients admitted following an abortion already had one (33%), two (15%) or three (23%) children (Shukri et al., 2015). Similarly, a study conducted in Kinshasa about pregnant women concluded that if abortion is used by the younger generation to delay the start of reproduction, it is also use by older women to space births (Svanemyr&Sundby, 2007).

Household wealth.Several studies have shown the correlation between abortion and the household wealth of women.The abortion practice is associated with household wealth of women. In fact, the higher the household wealth increases, the practice is high. In Egypt, forexample, richwomen in urban areas are more likely to practice abortions than poor women (Mafuta&Beya, 2006).

Employment status.Studies have shown that the abortion rate is particularly high among girls attending school and the desire to stay in school is one of the main reasons mentioned by African women to justify induced abortions (Maina, 2015). Other studies have also demonstrated a positive link between employment status and the likelihood of an abortion (Adjei, 2015). But according to Izugbara, et al. (2009), loss of job, unemployment and povertyrepresent mayor reasons to terminate a pregnancy.

The level of education.According to the literature, there is a positive and strong relationship between contraceptive use and the level of education of women. It appears that there is a variation for the recourse to abortion among educated women (Adjei, 2015). In Gabon, the proportion of women who resort to abortion increases with the level of education. Thus, women with secondary education (19%) are three times likely to practice abortion compared to those with no education (7%). (Mendame, 2015). According to Adjei, et al. (2015) the more a woman is educated the more she is likely to resort to pregnancy termination.

One study showed that women who were not using modern contraceptive methods had often resorted to traditional methods (Mendame, 2005). In Ethiopia, 83% of women did not use contraception before pregnancy (William A. 2004). In Togo, only 19% of women used contraception before pregnancy (Djoke, 2004). In Mali, the occurrence of unwanted pregnancy among women using a contraceptive method before the abortion is explained in 39% of cases in failure (William, 2004). According to Adjei (2015), theprevalence of contraceptive use and the presence of family planning services reduce the rate of abortion. In Egypt, the main determinant of unsafe abortion is lack of access to contraceptive methods and services and contraceptive method failure (Ezzeldin et al, 2010).

The goal of this study is to determine the factors associated with the pregnancy termination in Cote d'Ivoire.The study is based on a quantitative approach. The DHS-MICS 2011-2012 is a survey based on quantitative data. Thus, in order to achieve this study project, a quantitative approach was used to study the problem.

The design used tostudy the problem of this research is a cross-sectional study from Demographic Health Survey (DHS) Macro Database. Data was collected from December 2011 to march 2012 and was used to provide demographic and health phenomena and also AIDS and malaria indicators to managers and administrators of population and health programs in Côte d'Ivoire. The survey used closed-ended questions. Data collection focused onwomen demographic, socio cultural, socio-economic characteristic. In addition, the survey examined reproductive health behavior, contraception, fertility preferences, knowledge and prevalence of self-reported sexually transmitted infections reproductive health, reproductive background,abortion, pregnancy termination rate, prevalence of contraceptive methods, HIV/AIDS and Malaria.

Arrangements for the study. The DHS-MICS-11/12 survey was conducted from December 2011 to March 2012 as a part of the international program of Demographic and Health Survey (Demographic and health Surveys-DHS). But contrary to the previous surveys (DHS), this survey is a combination of DHS (Demographic and Health Survey) and MICS (Multiple Indicator Cluster Survey).

The DHS-MICS was conducted by the Ministry of Health and the Fight against AIDS (MSLS) in collaboration with the National Institute of Statistics (INS) for the implementation of the survey. The DHS-MICS-11/12 has received technical assistance from ICF international and with technical and financial assistance of the United States Agency for International Development (USAID), United Nations Children's Fund(UNICEF), United Nations Population Fund (UNFPA), the European Union, the World Bank, the Global Fund and The Joint United Nations Program on HIV and AIDS (UNAIDS). The survey was funded by the government of Côte d'Ivoire. But what we can say about the arrangement for the study, is that, all the protocols for hemoglobin testing, HIV and malaria were approved by the National Ethics Committee for Life Sciences and Healthcare (CNESVS), by the Ethics Committee (Institutional Review Board) of ICF International, and by the Committee Ethics of CDC (Center for Disease Control and Prevention) in Atlanta (USA) (DHS-MICS,11/12).

Participants. In this section we will present the survey coverage including the sampling and the participants. The DHS-MICS-11/12 (EDSCI-III) survey focused on the population of individuals residing in ordinary households for the whole country. The sampling of EDSCI-III is based on a survey by stratified two-stage cluster. For the first stage, a total of 352 clusters were randomly drawn from the list of clusters of AISCI-2005. A total of 161 clusters were drawn in urban place of residency and 191 from rural zones were selected by taking a systematic sampling probability proportional to the size. The size of the census districts is the number of households. (DHS-MICS-11/12, 2012). A total of 10,413 Households were selected and among them 9,873 households were identified during the survey. Of these households 9873, it is 9686 have been successfully surveyed, representing a response rate of 98%. The household response rate is higher in rural zone (99%) than urban areas (97%) (DHS-MICS-11/12, 2012). In the 9,686 households surveyed, 10,848 women aged 15-49 were identified as eligible for the individual survey, and the interview was conducted successfully with 10,060 of them (response rate of 93%). For thesurvey among males, a total of 5,677 were eligible and 5135 were interviewed (response rate of 91%) (DHS-MICS-11/12, 2012). The response rates observed in rural and urban areas are higher for women (95% against 91%) than those obtained for men (94% against 86%) (DHS-MICS-11/12, 2012).

Individual questionnaire of women. The household questionnaire used to establish the eligibility of persons to be interviewed individually. The individual questionnaire of women was used to record information about theeligible women and women of childbearing age (15-49). It also determines the reference population for the calculation of certain demographic rates and indicators (DHS-MICS-11/12, 2012). The questionnaire includes the main sections below.

The section on demographics consists of information including the age, the place of residence, date of birth, sex, education, literacy, region, exposure media and the birth rate.

The section on family planning contains information on knowledge of contraceptive methods, the use of contraception, source of supply of contraception, the future use of contraception, for family planning. The section on sexually transmitted diseases continues the information on STIs in general, with particular emphasis on knowledge of STI / AIDS risk perception of AIDS awareness and condom use. The section of health of mother contains information on prenatal care, tetanus vaccination, pregnancy outcome, childbirth, postnatal care and perceived problems for access to health care for women. (DHS-MICS-11/12, 2012). For analysis, the women individual questionnaire will be used for the factors associated to the pregnancy termination.

In this section we present the investigators training and data collection in order to show all the process of training and data collection. Data collection was done by well-trained surveyors.

Training. All collection procedures EDS-MICS 2011-2012 have been pre-tested. Twenty (20)surveyors were recruited and trained for four weeks on the filling and the structure of questionnaires and test procedures. Training was divided in a theoretical phase and a practical phase. Pre-test activities of the field were achieved in a precarious and populous neighborhood, a residential and a village of Abidjan and its surroundings (DHS-MICS, 11/12). The lessons learned from this pilot survey were used for the finalization of the survey documents.The National Institute of Statistics (INS) has recruited 144 candidates who have received training during one (1) month. The women interviewers were responsible for interviews with women while men investigators were in charge for interviews with men. (DHS-MICS,11/12).

Data collection. Data collection started on December 5, 2011 to completely ended on May 11, 2012. The different teams were deployed in their respective areas of work all over the country, according to their language skills to facilitate communication.

For data analysis, we will use two approaches: a descriptive analysis and an explanatory analysis(multivariable). These analysis methods will allow us to identify the factors associated with the pregnancy termination.

Bivariateanalysis. At the bivariate level we will show the associations between pregnancy termination and the different independent variables. With Statistics Chi2, we will evaluate the significance of the various associations.

Multivariableanalysis. As the relationships that will be highlighted at the descriptive analysis can be misleading, we will resort to multivariable explanatory models to identify net effects of each variable and mechanisms of action of various factors to the explanation of pregnancy termination. Since the dependent variable, pregnancy termination, is binary, we will use the multivariable logistic regression models. Logistic regression provides among other statistical interpretations of the results:

In this section we will present the descriptive analysis and multivariable analysis through regression logistic (step by step models).Before analyzing the data, it is important present the data in order to see the representativeness of variables. Then, in this section we presentthe distribution of variables. (Table1), the descriptive analysis and multivariate analysis (regression logistic with the step by step model) in order to identify the factors associated with the pregnancy termination. Pregnancy termination variable includes spontaneous abortion or induced abortion or miscarriage.

In this section we present the distribution of the different variables of the study (Table 1).

Table1 shows a total of 10,060 participants (women). The mean age of the respondent at time of survey was 28.52 years (standard error=0.092) showing that an important part of the participants is young.For the women of childbearing age, 39.52% of participants are aged of 15-24 years. Women of 25-34 years old represent 39.52% and those who are 35-49 years old represent 26.38%. The majority of participants (57.10%) reported having no education level, 23.33% reported completion of primary education while 19.57% reported secondary education and plus as education attainment. For the religion variable, 40.20% of participants are Muslims, 45.69% are Christians and 14.11% represent the animists (traditional religions).The survey results show that 64.15% of participants were married,while single women was35.85% of the participants. For theplace of residency, 51.39% ofparticipants reportedliving in urban area and 48.61% residing in rural areas. The survey results show that 29.29% of participants are working and 70.69% are not working. For the household wealth, 36.66% arepoor, while 20.87% are in middle situation andrich participants represent 42.74% of the sample.The percentage of women who don't use contraceptive method is high. Indeed, there are 81.1% of the participants who does not use a contraceptive method, while 18.79% use a contraceptive method. Women without children represent 25.47% of participants in this sample and almost half (49.79%) of participants have 1-4 children while those who have 5-9 children represent 22.55% and 2.19% have 10-15 children. For the female circumcision (female genital mutilation) variable, 56.01% of women are circumcised while 43.99% of participants are not circumcised.For the pregnancy termination, 82.07 % of participants reported that they have never had a pregnancy termination while 18.03% have ever had a pregnancy terminated.

Variables

Frequency (N)

Percentage (%)

Women of childbearing age

15-24

25-34

35-49

3984

3340

2736

39.52

34.10

26.38

Religion

Muslim

Christian

Animist

4044

4597

1420

40.20

45.69

14.11

Marital status

Married

Single

3607

6453

35.85

64.15

Place of residence

Urban

Rural

5170

4890

51.39

48.61

Education level

No education

Primary

Secondary&+

5744

2347

1969

57.10

23.33

19.57

Occupation

Working

No working

2946

7114

29.28

70.72

Wealth level

Poorest

Middle

Rich

3688

2100

4272

36.66

20.87

42.47

Contraceptive method

Yes

8074

1986

81.21

18.79

Total children ever born

0 children

1-4 children

5-9 children

10-15 children

2562

5009

2269

220

25.47

49.79

22.55

2.19

Female circumcision

Yes

5292

4157

56.01

43.99

Pregnancy termination

Yes

8217

1843

81.07

18.93

Total

10060

100

In this section we present the bivariateanalysis in order to identify the associations between the independent variable and pregnancy.

Proportion of women of childbearing age, 15 to 49, who reported that they have had a pregnancy termination

Variables

Frequency (n)

Proportion (%) of Women of childbearing age, 15 to 49, who reported that they have had a pregnancy termination

Probability of CHI2

Women of childbearing age

15-24

25-34

35-49

3984

3340

2736

11.07

20.60

26.10

0.000***

Religion

Muslim

Christian

Animist

4044

4597

1420

15.63

21.36

13.67

0.000***

Matrimonial status

Married

Single

6453

3607

20.72

14.03

0.000***

Place of residence

Urban

Rural

5170

4890

19.22

17.57

0.000***

Education level

No education

Primary

Secondary

5744

2347

1969

16.47

20.92

20.62

0.000***

Occupation

Working

No working

2946

7114

19.81

14.73

0.000***

Household Wealth

Poor

Middle

Rich

1727

1780

1910

16.81

16.62

20.46

0.000***

Contraceptive methods use

Yes

8074

1986

17.20

23.17

0.000***

Total children ever born

0 children

1-4 children

5-9 children

10-15 children

2562

5009

2269

220

11.36

19.50

23.49

19.09

0.000***

Female Circumcision

Yes

5292

4157

19.33

18.31

0.206

All the variables are associated to the pregnancy terminations at 1%, except female circumcision which is not significant (p=0.206). (See table2). There is a significant relationshipbetween age and pregnancy termination (p=0.000) at 1%. Women aged of 34-49 years are more likely to have pregnancy termination witha proportion of 26.10%. Women aged of 25-34 years represent 20.60% who have terminated a pregnancy while women aged of 15-24 yearshave a proportion of 11.07%. It appeared that the odds ratio of pregnancy termination increase with age.As shown in table 2, there is a significantrelationship between pregnancy termination and religion at 1% (p=0.000). Christian women have the highest proportion of termination pregnancy with 21.36%. Muslims have a proportion of 15.63% while the animistshave a proportion of 13.67% of pregnancy termination. The association between pregnancy termination and matrimonial status is significant at 1% (p=0.000). Single women are less involved in pregnancy termination (14.03than married women (20.72%) %). In addition, the place of residency is significantly associated to pregnancy termination at the level of 1%. Indeed, 20.72 % of women living in urban areas are more involved with pregnancy termination than participants living in rural areas(14.03%).The association between education and pregnancy termination has a strong relationship at1% (p=0.000). Women with no education level practice less pregnancy termination (16.47%) than women of primary education level (20.92%) and women of secondary and plus education level (20.62%). Also, the occupation is associated with pregnancy termination at 1%. In fact, women who are working are 19.82% to practice pregnancy termination while women who don't work (14.73%). In addition, there is a strong association with household wealth at a level of 1% (p=0.000). Indeed, rich women practicemore pregnancy termination (20.46%) than poor women (16.81%) and women who have a middle householdwealth (16.62%).

In addition, there is a significant strong relationship betweencontraceptive method useand pregnancy (p=0.000). In fact, 23% of women using contraceptive method practice pregnancy termination while women who do not use it represent 17.20% of participants. In addition, pregnancy termination and number of living children of women are significantly associated at 1% (p=0.000). Indeed, more womenhavelots of children they are likely to have a terminated pregnancy. Women with 5-9 children (23.43%) have practiced pregnancy terminationwhile 19.09% of women with 10-15 children and 19.4% of those with 1-4 children have had a pregnancy termination. But, only 11.36% of women without children have practiced pregnancy termination. However, male circumcision is not significantly associated with pregnancy termination (p=0.206). It appeared that female circumcision does not influence pregnancy termination.

For the bivariateanalysis, we sought to highlight the statistical associations between the independent variables and the pregnancy termination (dependent variable). The association tests have overall shown strong relationship between every variable and pregnancy termination.To reach the purpose of the study it is necessary to find out the factors associated to the pregnancy termination through multivariate analysis.

MultivariableAnalysis

The bivariateanalysis allowed us to identify the associations between the independent variables and pregnancy termination and mechanisms by which they influence this event.

Explanation of the step by step model (Table3).In table3, there is the bivariate analysis (unadjusted) between each independent variable and the pregnancy termination. In addition, there are the different models (Mode1, model2, model3) which are adjusted. The step by step model uses logisticregression. At every step, it is introduced a setor group of variables in order to see how these variables(or a variable) are going to affect the other variables in terms of odds ratio and test significance (á). The final model is Model3 which is used to determine the net effect of the factors influencing the pregnancy termination. We will use the different variations (odds ratios, á significance level) the models to determine the mechanisms by which the different factors (variables) influence the pregnancy termination.

Results from logistic regression for factors associated with the pregnancy termination among sexually active women, using DHS 2012 data.

Variables

Bivariate analysis(unadjusted)

Model 1

A to D

Model 2

A to G

Model 3

A to J

A. Women of childbearing age

15-24

25-34

35-49

***

Ref

2.08***

2.84***

Ref

1.92***

2.64***

Ref

1.74***

2.48***

Ref

1.76***

2.52***

B. Place of residence

Urban

Rural

1.12

Ref

1.34

Ref

0.96

Ref

0.94

Ref

C. Religion

Muslim

Christian

Animist

0.68***

Ref

0.77***

0.63***

Ref

0.77***

0.71***

Ref

0.85**

0.66***

Ref

0.84**

D. Marital status

Married

Single

***

1.60***

Ref

1.33***

Ref

1.42***

Ref

1.29***

Ref

E. Education level

No education

Primary

Secondary & plus

***

Ref

1.34***

1.31***

Ref

1.32***

1.57***

Ref

1.31***

1.55***

F. Occupation

No Working

working

***

0.70***

Ref

0.86**

Ref

0.88

Ref

G. Household wealth level

Poor

Middle (medium)

Rich

***

1.01

Ref

1.29*

0.90

Ref

1.26***

0.91

Ref

1.26***

H. Using contraceptive methods

Yes

***

Ref

1.08***

Ref

0.80***

I. number of living children

0 children

1-4 children

5-9 children

10-15 children

***

0.53***

Ref

1.27***

0.97

0.75***

Ref

0.94

0.75

J. Female Circumcision

Yes

1.07

Ref

0.89

Ref

Significance (model)

***

Pseudo R² (%)

8.66

9.58

10.1

All the three models are statistically significant at 1%. The Pseudo R² for model1, model2 and model3 are respectively 8.66%, 9.85% and 10.1%.Overall, (model3), the different variables influencing the pregnancy termination are women of childbearing age, religion, marital status, household wealth, contraceptive methods, number of living children.(Table3).

Women aged of 25-34 years and 35-49 yearsare respectively at 1.76 and 2.52 times more likely to perform pregnancy termination than those who were 15-24 years old.

The Muslim and animist women are 0.66 and 0.84 times less likely to resort to abortion compared to Christians. It appeared that married women are more likely to have had a pregnancy termination. Indeed, in the bivariate analysis, single women are 1.60 times more likely to have had pregnancy termination than those who are single. In the model3,married women are 1.29 times more likely to have had a pregnancy terminationcompared to single women.In the case of the education level, womenof primary and secondary and plus education level are respectively1.31 and 1.57 times more likely to have pregnancy termination compared to those without no education level. In addition, forthe householdwealth variable, rich women are 1.26 times more likely to have pregnancy termination compared to poor women. Moreover, women who do not use contraceptive method are 0.80 times less likely to have had a pregnancy termination compared to those who use it. In addition, women without children are 0.75 less times to practice pregnancy termination than those who have 1-4 children.

When we look at the different multivariate models, in the model2, after the introduction of education level, occupation and wealth index, it is noticed that, the religion variable with the category Muslim, which is significant at 1 % pass from odds ratio of 0.71 to 0.66. Single women oddsratio is1.29 in model1 increases in the model2 to decrease in the model3 to 1.29. with a decrease of 10%, but this variable remains significant. In the model3 after the introduction of the contraceptive methods, number of living children and female circumcision variables, it appeared that women of childbearing age,religion,marital status, education level, occupation (employment status)and householdwealth are still significant. In the final model, the place of residence, occupation,and female circumcision are not significant(model 3).

According to the data collected and the analysis, it is important to state about the study's hypotheses. The followinghypotheses were failed to be rejected:

Hypothesis #1(H1), which stated that the woman of childbearing age, marital status and religion explain the recourse to pregnancy termination in Cote d'Ivoire is not rejected.Results stated that woman of childbearing age, marital status and religion have an effect on the practice ofpregnancy termination.

Hypothesis #2(H2),which stated that the level of education and the household wealth influence pregnancy termination in Cote d'Ivoirefailed to be rejected.Results stated that education level influences directly pregnancy termination. In fact, women of primary and secondary and plus education level are more likely to practice pregnancy termination compared to women without no education level. Also, household wealth has an effect on pregnancy termination. Rich women are more likely to have a pregnancy termination compared to poor.

Hypothesis #3(H3), which stated that contraceptive use explains pregnancy termination failed to be rejected. Results stated that women using contraceptive methods are more likely to resort to pregnancy termination than those who don't use contraceptive methods.

The main objective of this research project was to identify the chain of factors behind pregnancy termination in Cote d'Ivoire. The specific objectives for this study were to identify the profile of women of childbearing age (15-49) exposed to pregnancy termination and to identify factors associated with pregnancy termination and their mechanisms. The results reached may be explained by the combination of several reasons and relationships.

The present study suggests that the sociocultural (women of childbearing age, religion, marital status),socioeconomic (education, occupation, household wealth) and institutional (contraceptive method use) factors and others factors such as the number of living children of women influence pregnancy termination.

Women aged of 25-34 and 35-49 years are more likely to terminate a pregnancy compared young women aged of 15-24 years. That situation can be explained by the fact that those women want to reduce their number of children. (Mendame, 2000). Thus, pregnancy termination (induced abortion) continues to be used as a family planning by women. (Adjei, 2015). There is also the stigmatization of induced abortion premarital pregnancy which is not allowed by almost of societies. Several studies have found the same results. (Mendame, 2005). Thereis also the lack of finance. According to some researchers (Adjei,2015; Svanemyr andSundby, 2007), the lack of money isa reason reported by young couples to terminate apregnancy. Looking at the model2, the introduction of economic factors (education level, occupation and household wealth) maintains the significance of women reproductive age. Indeed, the influence of the women of childbearing age on pregnancy termination passes through the socioeconomic factors (education, occupation and household wealth).

The religion has an influence on termination of pregnancy. In fact, it appears that religion is a determinant factor of pregnancy termination. The Muslim and animist religions are less likely to practice pregnancy termination compared to women of Christian religion.Indeed, for Muslims, a woman who aborts can be rejected by the community and the family. The Christian religion may tolerate the pregnancy termination such as induced abortion than the others religions because this religion refuse social exclusion (Izugbara, 2009).The stigma of a premarital pregnancy is severely punished. (Adjei, 2015). This finding is similar to a study achieved in Ghana where abortion is seen like a taboo topic and a shameful act. (Adjei, 2015). Regarding the matrimonial status, the results showed that the marital status is significantly associated with the pregnancy termination (Table3). Indeed, married women are more likely to practice pregnancy termination than single women. This finding is consistent with many otherresearch results (Izugbara et al, 2009). In fact, married women terminate a pregnancy in order todelay a pregnancy or to limit the number of children or to avoid unintended pregnancy. (Izugbara et al, 2009). Also,there is a needof contraceptionto limit births. In fact, in Cote d'Ivoire 15 % of familial planning need is not satisfied (UNICEF, 2013). There is also the fact that many pregnancies are undesired. Thus, married women prefer to terminate a pregnancy. (Adjei, 2015). Economic reasons (lack of money to meet the needs of a child) can be evoked to justify this situation. (Svanemyr&Sundby, 2007). We can also mention the stigma of a premarital pregnancy, an unstable relationship with a partner. There is also the fact that some women think that they are too young to have a child. There is also the lack of contraceptive methods and services may represent causes for single women to terminate a pregnancy. This result is consistent with research achieved by Izugbara et al, (2009). In this study, the intention to reduce the number of children is cited by women as the reason to abort (Izugbara et al, 2009). In addition, the finding showed that the education level is significantly associated with the pregnancy termination. Moreover, it is observed that when education level of women is higher, the more they are more likely to have pregnancy termination (Mendame 2000). The influence of the education level is reinforced with the control of cultural and economic variables in the model2 and model3. In fact, in the model3, the odds ratios of the variable remain significant. In addition, women of primary or secondary and plus level are likely to practice pregnancy termination compared to women than women who are not educated. In fact, educated women live in urban areas. Urban areas have influence onwomen behavior factor with regard to sexuality. Many educated women are more focused on their professional career before having a child. Also, because of the fact women whoare students and are not married prefer practicing pregnancy termination in order to fulfill their scholarship or before they get married. Educated women have good knowledge and attitude in favor to the contraceptive methods, but the fact that they cannot satisfy the unmetneed for contraception methods needs push them to practicepregnancy termination in order to avoid an unintended pregnancy.

In addition,women without work are less likely to practice pregnancy termination compared to those who work. According toIzugbara et al. (2009) loss of jobs and poverty. Then in consequent, those women have the desire to limit their number of children or want to delay a pregnancy. Also, this situation can be explained by the lackof avaibility of contraceptive methods and family planning services. Looking the at the employment status (occupation), this variable is significant at 5% level in the model2, but becomes not significant in the model3 after the introduction of contraceptive method, number of living children women and female circumcision variables. These results show the action of work status passed through this group of variables.Also, the household wealth status variable has a significant effect on pregnancy termination.In this study, it is observed that richest women are more likely to practiced pregnancy termination than poor women. This finding is similar with several other studies in Africa. Indeed, according to Adjei et al.(2015) wealthiest are found to be more likely to have an induced abortion.Women who are rich are likely to practice pregnancy termination. (Adjei et al, 2015). In model2 and model3, this variable remained significant. Controlling this variable with the occupation, most of rich women live in urban area, wherethey have the opportunity to get family planning services and obstetric care but the unavailabilityof contraceptive methods and services may push them to practice apregnancy termination in order to avoid an unintended pregnancy. Thus, they have the chance to get a healthcare assistant for the obstetric care. Those women have access to obstetric care but have unmet needs for familial planning. Thus, all these situations push them propensity to practice abortion to avoid undesired pregnancy.

Moreover,the effect of the contraceptive method remains significant from the bivariate analysis to the final model (model3). Women who don't use contraceptive method have a high inclination to pregnancy termination compared to women who use contraceptive method. Those women don't have enough knowledge of contraceptive methods. This variable has adirect actionon pregnancy termination.In fact,contraceptive methods are used by woman to avoid unintended pregnancy or to limit their children. This variable is very relevant sincebecause its influence on the model (model3) is important after the control of the sociocultural and socioeconomic variables and the other variables.Failure of contraceptive method can explain the fact that women using contraceptive method are more likely to practice pregnancy termination. Thus, women practice a pregnancy termination when a contraceptive method fails (Adjehi, 2015). This result suggests that, women used contraception method as an option as a family planning for unplanned pregnancy in the case of a contraceptive method failure. (Musalu, 2006).In addition, pregnancy termination is significantly influenced by the number of children a woman already has. The number of living children of a woman significantly influences pregnancy termination. Indeed, woman without children have less propensity to practice termination pregnancy. The action of this variable passes through contraceptive use method. Thus,women want to limit their numberof children by using contraceptive methods. Thus, when this method failed and these women have unmet need for contraception pushed them to terminate their pregnancy.

However, contrary to the other variables, the place of residency, occupation and female circumcision variables are significantly associated with the pregnancytermination. However, this study presents few limitations such as the combination of spontaneous, induced abortion and miscarriage into one variable (pregnancy termination). Also, there are problems of abortions and miscarriage estimations.

Pregnancy termination is a serious public health problem that the governments of developing countries want to fight with the support of international organizations.Few studies have examined the phenomenon, due to lack of data related to its illegality such as induced abortion. Among those available, very few have actually examined factors that might find factors associated with this practice. Our goal was to identify the factors associated with pregnancy termination.For the review of the literature section we are interested in different explanatory approaches to pregnancy termination in Africa, to the evolution of legislation and ideological debates. A review of the literature guided us to identify the different variables that may determine pregnancy termination. We were able to identify later the specific assumptions. Thus, this study generally assumes that pregnancy termination depends on socioculturalsocioeconomic factors, contraceptive methodsuse, number of living children. The methodology section presented the data sources used for the study and theanalysis method. The data used for this study are from the first Demographic and HealthSurvey of Cote d'Ivoire achieved in 2012.It was conducted in two phases for the analysis of data: first a bivariate analysis and then a multivariable analysis.

The main results of the descriptive analysis showed that most ofindependent variables (age, religion, marital status, education level, occupation, household wealth, number of living children and contraception use) are statisticallyassociated with pregnancy termination. Overall, the relationship test showed a strong correlation between the independent variables and pregnancy termination.Logistic regression models were used to highlight the influence of socio-economic, socio-cultural and institutional factors on pregnancy termination. Regarding the explanatory analysis, we used logistic regression, allowing us to identify the factors influencing pregnancy termination in Cote d'Ivoire.

After the analyses, it appears thatpregnancy termination is under the combined action of sociocultural, socioeconomics and institutional factors.

- For the socio-cultural factor, the results show that it is the women reproductive age, religion and marital status that influence thepregnancy termination.

- For socioeconomicfactors, it appears that, education level, occupation and household wealth are the main factors associated with the practice of pregnancy termination in Cote d'Ivoire

- Regarding the institutional factors, the findings show that the contraceptive methods women have an effect on the pregnancy termination. And the other factors such as number of living children influence the pregnancy termination.

The logistic regression models brought out the mechanisms behind pregnancy termination.Looking at the study results reached it is necessary to make the following recommendations.

Based on the results of this research project and to address the burdenof unsafepregnancytermination, the following recommendations are provided for consideration:

implementation recommendations, recommendationstoimprove the research studyand recommendations for future study.

of abortion, the Ministry of Public Health should increase the number ofqualified medical personnel for the treatment of pregnancy termination and specially abortions.

2. Improving access to induced abortions through intensified educational programs on the current abortion law and its provisions.

3. Ameliorate the rights of persons necessitating abortion services in Cote d'Ivoire.

4. Local, national and international organizations could use the results of this study to intensify the information and education of the population about the dangers of abortions and associating religious authorities in the programs.

5. Results of the study should be shared with the Department Public Health in order todevelop plans of action that address specific issues related to pregnancy termination.

6. Department of Health and International Organization should make contraceptive methods accessible and affordable for women

1. The study is at community level but in order to get a good estimation of pregnancytermination, it is crucial to add hospital-based (public and private) datato the data collected to better understand the main risk factors that can helpto address specific problems.

2. The section of pregnancy termination must be well documented in the questionnaire in order to get accurate informations and data.

1. Researchers must clarifyother factors that can be associated with the pregnancy termination such as psychosocial factors.

2. In order to provide a better estimation of pregnancy termination it is recommended to achieve a specific survey on spontaneous, induced abortions and miscarriages;

3. There is a need for qualitative researches to understand better the key risk factors for induced abortion to help plan specific interventions aimed at eliminating abortion-related health risks.

4. Researchers shall use data from a nationally- representative sample of public and private health facilities of medical records.

Adjei, G., Enuameh Y., Poku K. A., Baiden F., Nettey O. E., Abubakari S., ....Gyaase S. &Owusu A.S. (2015). Predictors of abortions in rural Ghana: a cross-sectional study. BioMedCentralPublic Health, 15,202.

Barrere, M. (2001). Avortement, Enquête Démographique et de Santé, Gabon 2000, 86-98.

Bazira, E.R., (1992). Induced abortion at Mulago Hospital Kampala,1983-1987: A case for contraception and abortion laws' reform. Tropical Health, 11(1), 13-16.

Bajos, N., Ferrand M, (2002). De la contraception à l'avortement. Sociologie des grossesses non prévues. Paris: Inserme,345.

Borgatta, L. (2014). Labor induction termination of pregnancy. Global Library of Women's Medicine,10444.

Bibang, F. (2010). Recours aux methodes modernes d'avortement prologue au Gabon, recherche des facteurs. Cahier de l'IFORD, 56-70.

Calves, A. E. (2002). Abortion risk and decision making among young people in urban Cameroon. Studies in Family Planning, 33(3), 249-260.

Dodo, M. (2003). Risky sexual behavior in Cote d'Ivoire. (Unpublished master dissertation).Institute of Training and Demographic Research, Yaoundé (Cameroon)

Fawcus,S. R., &FrecogM. (2010). Best Practice &Research Clinical Obstetrics and Gynaecology. 22, 3, 533-548.

Goyaux, N. F., Yace S. C., Welffens-E., Thonneau P. (1999). Abortion complications in Abidjan, (Ivory Coast). Contraception 60, 107-109.

Goyaux, N., Alihonou E., Diadhiou F, Leke R., Thonneau P.F. (2001). Complications of induced abortion and miscarriage in three African countries: Ahospital-based study among WHO collaborating centers. Acta ObstetriciaGynecologicaScandinavica, 80(6), 568-73.

Guillaume,A, Desgrees L. A., Koffi N.G., Zanou B., (1999). Le recours à l'avortement: La situation en Côte-d'Ivoire. ENSEA, IRD, Abidjan, 50.

Guillaume, A., (2003). Le rôle de l'avortementdans la transition de la fécondité à Abidjanaucours des années 1990. Population Studies, 15, 58-63.

Guillaume,A., 2000, Abortion in Africa: A birth control method and a public health issue. In Ceped News, (8), 1-4.

Guttmacher Institute, (2015). Facts on abortions in Africa. Retrievedfrom https://www.guttmacher.org/.../IB_AWW-Africa.pdf

Haddad, L. B., & Nour N. M. (2009). Unsafe abortion: Unnecessary maternal mortality. Reviews in Obstetrics andGynecology, 2(2),122-126. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PM709326

Izugbara, C. O., Otsola K. J., & C. Alex E. (2009).Men, women, and abortion in central Kenya: A study of lay narratives. Medical Anthropology, 28(4),397-425.

Jakubowicz, D. J., JakubowiczM. J, S., Roberts K. A., Nestler J.E. (2002). Effects of Metformin on early pregnancy loss in the Polycystic Ovary Syndrome. Journal of Clinical Endocrinology & Metabolism, 87(2),524-529.

Kasolo, J. (2000). Abortion in Uganda.Initiatives in Reproductive Health Policy, (2), 9-10.

Konate M., (1999). Les conséquences sociales de l'avortement provoqué à Bamako. CILSS, INSAH, CERPOD, Bamako, 91.

Madebo, T., Tadie G.T., (1993). A six-month prospective study on different aspect of abortion. Ethiopian Medical Journal, 31(3), 165-172.

Menachem, F. I., Ido S.&Yechiel Z. B. (2015). Pregnancy prevention and termination of pregnancy in adolescence: Facts, ethics, law and politics. Imaj, 17,120-126.

Mafuta, E., Musalu, P. & Mutombo, B. (2006). Impact of socio-economic factors in the practice of unsafe induced abortions in democratic republic of Congo: Case of women population of Kikwit(Unpublished thesis). College of Public Health,University of Kinshasa.

Magnani, R. J., Rutenberg N., McCann G. H. (1996). Detecting induced abortions from reports of pregnancy terminations in DHS calendardata. Studies in Family Planning, 27(1),36-43.

Maina W. B. (2015). Factors associated with repeat induced abortion in Kenya. BioMed Central Public Health, 15,1048.

Mendame, M. W., (2000). Determinants of induced abortion in Gabon(Unpublished master dissertation).Institute of Training and Demographic Research, Yaoundé (Cameroon).

Moyabi, S. (2002). Les determinants de la fecondite des adolescentes en Cote d'Ivoire(Unpublished master dissertation).Institute of Training and Demographic Research, Yaoundé (Cameroon).

Ngwe, E. (2005). Connaissances, attitudes et pratiques relatives à l'avortement à Yaoundé et Douala. CCRA, ALVF, 86.

Rwenge M. (2002). Culture, sexual behavior and HIV/AIDS. Les Cahiers de l'IFORD, 8(2)198-201.

Shukri, F.M., Chimaraoke I., Moore A. M., Mutua M., Kiman E. W., Abdhalah K. Z., AkinrinolaB.,Egesa C. (2015). The estimated incidence of induced abortion in Kenya: a cross-sectional study. BioMed Central, 15, 185.

Davis,F.A.(2011). Abortion.In Taber's Medical Dictionary: Taber's Cyclopedic Medical Dictionary.

Svanemyr, J. &Sundby J. (2007). The social context of induced abortions among young couples in Côte d'Ivoire. African Journal of Reproductive Health, 11, 2, 13-23. Retrieved from http://www.jstor.org/stable/25549712

The United Nations Children's Fund (UNICEF).(2015). La situationdes enfantsdans le monde 2015 - Résumé.Retrieved from http://www.unicef.org/infobycountry/cotedivoire_statistics.html

The Johns Hopkins University School of Medicine. (2012.) The Johns Hopkins manual of gynecology and obstetrics(4 ed.).Lippincott Williams & Wilkins.

Varga C.A.(2002). Pregnancy termination among South African adolescents.Study in Family Planning. 33(4), 283-298.

World Health Organization. (2000). Rapport sur la santé dans le monde, 2000 - Pour un système de santé plus performants. Retrieved from http://www.who.int/whr/2000/fr/

World Health Organization. (2011). Unsafe abortion: Global and regional estimates of the incidence of unsafe abortion and associated mortality in 2008(6^th ed.). Geneva.