Sexual coercion at first sexual debut and associated factors among young females in rural areas of Boset District, Eastern Ethiopia: A mixed-method study

Objectives: Assess the prevalence and associated factors of sexual coercion at first sexual debut among young females in rural areas of Boset district, Oromia Region, Ethiopia. Methods: Cross-sectional community-based study was conducted using qualitative and quantitative data collection methods. A sample of 632 young females aged 10-24 years was taken from six rural Kebeles by systematic random sampling for quantitative and data was collected using a pre-tested structured questionnaire. The collected data was analyzed using SPSS version 23. Principally Binary Logistic regression model was fitted. Moreover, eight focus group discussions targeting different groups were held and analyzed thematically. Results: Prevalence of sexual coercion at first sexual debut was 36.5% (95%CI: 34.6%-38.4%) among sexually active respondents. Married young female (AOR,0.71;95%CI: 0.42,0.81), living away from their parents (AOR,5.07;95%CI:1.07,24.01), age group of 15-19 (AOR,0.19; 95%CI:0.06,0.54), alcohol consumption (AOR,2.44;95%CI:1.17, 5.08) and Khat chewing (AOR, 8.30;95%CI:1.89, 36.38) were factors predicting the likelihood of having sexual coercion at sexual debut. Conclusion: Sexual coercion at fist sexual debut is a major public health problem among young females in the study community. The emerging program interventions need to take this problem into consideration.


Introduction and background
Sexual coercion is the act of being physically, psychologically, fi nancially, or otherwise forced or tricked into engaging in sexual activity [1,2]. Most commonly the victims of sexual coercion are women and children [3,4]. These include acts of being worn down by someone who repeatedly asks for sex being lied to or being promised things that weren't true to trick you into having sex; having someone threaten to end a relationship or spread rumors about you if you don't have sex with them; and having an authority fi gure, like a boss, property manager use their infl uence or authority to pressure young female into having sex [5,6]. It is commonly believed that sexual coercion is perpetrated by the male sex against the female sex [5]. For example, in a community-based study, 33% of young females reported having sexual coercion from men, and 70% of them knew their perpetrator (a boyfriend and an acquaintance) [2].
Studies in the fi elds of sexual coercion at sexual debut, including rape are usually based on biased estimates to elicit the relationship between the victims and rapist characterization of the prevalence is challenging [2,7]. There are also popular myths about both victims and perpetrators [8]. Some of the commonly perceived myths are sexual coercion is performed by strangers in dangerous places; the offender or perpetrators are deviates or distorts, and they are more likely to socially and psychologically normal acquaintances [2,8].
Study examined sexual coercion at fi rst sexual debut among rural young females in low-to-middle income counties found 23% on average [9,10]. Modifi able predictors of sexual coercion at fi rst sexual debut are poorly understood, particularly among young people in rural areas who may have less access to sexual health resources [6,7,11]. In the context of sexual coercion drugs/alcohol consumption by both men and women is one of the most frequently cited to lower the victim's inhibition (loosen them up) or lessen their verbal resistance to sexual advances. Ages, parental communication, marital status, education, living arrangements are among many factors frequently associated with sexual coercion at sexual debut [4,[12][13][14].
Sexual coercion has been linked to a variety of negative outcomes such as psychiatric symptoms, poor school performance, and behavioral problems [4,15]; among adolescent girls, it has been associated with pregnancy and abortion [11]. Sexual coercion also may increase adolescents' risk for infection with Human Immunodefi ciency Virus (HIV) and other sexually transmitted infections [6].
However, there is a paucity of information on the prevalence and associated factors among young people in urban and university settings of Ethiopia [12,[16][17][18][19]. Also, in rural Ethiopia, where more than 80% of the young people live, the prevalence and associated factors of sexual coercion at fi rst sexual debut are not well explored. Obtaining reliable estimates of sexual coercion among rural adolescents is important given the challenging social norms and myths. The study assesses both prevalence and associated factors of sexual coercion at fi rst sexual debut and associated factors in the rural settings of Ethiopia during 2019.

Study area
The study was conducted in rural Kebeles (the lowest administrative unit in government structure) among young females (10-24 years) in Boset district, Eastern Ethiopia. The district is found in the East Shoa Zone of Oromia Region 120 kilometers away from Addis Ababa. The district has a total population of 497, 143 during in 2019/20 projected from the 2007 Census conducted by the Central Statistical Agency of Ethiopia, considering 2.9 % as a rate of natural increase [20]. The total number of those aged 10-24 years in the district was estimated to be 154, 114 based on the assumption of 31% of all segments of the population [20].

Study design and period
A community-based cross-sectional design was employed using both quantitative and qualitative data collection methods from May to June 2019.

Population
All young females (10-24 years) and those living in randomly selected households of rural Kebeles in Boset district were the source and study population, respectively. Those young females who have lived at least for six months in the district prior to the survey were included, while those critically ill and unable to respond to the questionnaire were excluded from the study.

Sample size
The required sample size was calculated by a single population proportion formula using 50% proportion of sexual coercion among young females in the absence of information from rural settings, at 95% confi dence level, the precision of 5%, design effect of 1.5, and adding 15 percent for the possible non-responses due to the sensitive nature of the study. Finally, a minimum of 662 study subjects was needed for the study.

Sampling procedure
A multistage sampling method was used to randomly select six from 40 rural Kebeles in the Boset district. To identify eligible, the household census and numbering were carried out prior to data collection. Based on the identifi ed households having the target group, the probability proportional to sample size allocation was carried out. Finally, systematic sampling was used to include the eligible into the study. Indeed, only one was selected by lottery method if the household has two or more eligible to control for potential intra-household correlation.
For the qualitative approach a Focus Group Discussions (FGDs), two per group were sampled and it was determined based on the saturation of information or redundancies

Data collection
Quantitative data were collected using a structured interviewer-administered questionnaire developed after reviewing relevant works of literature [7][8][9]11]. The questionnaire was translated into the regional working language (Afan Oromo) and translated back to English by experts to ensure its consistency. To ensure the validity of the questionnaire, pretest was carried out in a similar setting on young females who had similar socio-demographic characteristics by considering

Data analysis
The collected data were cleaned, coded, and entered into Descriptive statistics were used to determine the proportion of sexual coercion among the study participants. The associations between the study variables were computed using binary logistic regression analysis. In the fi nal model, variables having a P-value of < 0.25 at a bivariable level were chosen and entered into the fi nal model using the forward selection method to estimate the adjusted effects of the predictor variables. The odds ratio with 95% confi dence intervals was calculated to indicate the strengths of the associations between the study variables. The signifi cance of the association was declared at a p-value of 0.05. The multicollinearity between the independent variables was checked using the Variance Infl ation Factor (VIF).
Those variables having the value of 10 were excluded from the model. The model's ability to correctly classify the outcome was assessed using the Receiver Operating Characteristic curve (ROC). For the qualitative information verbatim transcriptions were used. Texts were imported to Open Code 2007 program to facilitate the coding process [21]. Relevant codes were arranged into categories to arrive at predetermined themes and thematic analysis was conducted. The trustworthiness of qualitative information was ensured by recruiting appropriate discussants and member checking was also used after write-up of the texts. The results of qualitative information were used to complement the quantitative results and discussions.

Ever experience sexual coercion at fi rst sexual debut:
Respondents were lebeled if they had unwanted or unwillingly completed penetration because of one or more of the followings: deception (promise) or reward, the threat of nonphysical punishment or verbal pressure, exchange of sex for money/ gifts/favor (transactional sex), by the use of physical force/rape and use of substances (alcohol, chat, and drugs) [11].
Khat: A stimulant plant leaf chewed by people mostly practiced in some social traditions in parts of the Middle East, such as Saudi Arabia and Yemen, and in Eastern Africa, such as Ethiopia and Somalia [11].

Sexual coercion:
Unwanted or unwillingly completed penetration because of one or more of the followings: deception (promise) or reward, the threat of non-physical punishment or verbal pressure, exchange of sex for money/ gifts/favor (transactional sex), by the use of physical force/rape and use of substances (alcohol, chat, and drugs) [11].
Sexual debut: Those young females experienced penetrative sexual intercourse for the fi rst time [11]. Young people: are those who are in the age group of 10 to 24 years.

Ethical considerations
As part of the staff research, the ethical clearance was received from an Ethical Committee and Institutional Review Board (IRB) of the Adama Hospital Medical College. The offi cial letter was issued to the concerned bodies by the Department of Population Health. Consents from the participants and or assents from parents or guardians of the study participant's age less than 18 years were secured. Before undertaking the interview and discussions, participants were informed to have a full right to agree or disagree and or withdraw at any time. All interviews and discussions took place in complete privacy and comfortable places.

Socio-demographic and behavioral Characteristics
It was planned to include 662 young females in the study and 632 were participated, making a response rate of 95%.
One hundred twenty-four (19.6%) were in the age range of 10-14 years. The mean (+SD) age of the respondents was 17.2 (+3.3) years. Most (83.7%) were not married (single), while 13.8% were ever married. About seven in ten (70.4) of them were students and 4% of them have never attended education and were considered to be illiterate during the data collection time. More than four in ten (41.9%) currently living with their parents, while 5.1% were living alone. About one of seven (14.7%) of the participants used alcoholic beverages and 2.7% chewed Khat ( Table 1).

Prevalence of sexual coercion at fi rst sexual debut
The proportion of young females who ever experienced penetrative sexual intercourse was 233 (36.9%). The mean (+SD) age at fi rst sexual commencement was 16.64 (+1.97)  Figure 1).
Regarding the types of sexual coercion at fi rst sexual debut, the victims faced multiple forms at the same time. More than a quarter (27.6%) were physically forced/raped, while nearly one in fi ve (21.3) were deceived or misled by promising words, and the rest 13.7%, 9.3%, 4%, and 0.9% of the victims started their fi rst sexual intercourse by verbal threats, exchange for gifts/ property, by making them drunk or after chewing Khat and watching pornographic movies/pictures, respectively ( Figure  2). The result is substantiated by the results of FGD explored usually the sexual intercourse with girls is being practiced at the study area on the basis of economic support, physical forces/ rape, and deception or misleading words coming from men for defl owering by repeatedly saying, "I love you very much".
One of the discussants said "Qalbiin debate amma ija xaafi i geese, isseeyyuu yoo dhungatani lafa keessii" meaning the discerning range of mind of females towards sexuality is as small as a seed of "millet", which even she can lose it if kissed (38 years married women). This implies that females are easily fl attered over sexual desire.
The discussants also mentioned pressure from peers is not uncommon. Men can make an agreement with their intimate friend of female that makes a mediating task in persuading another female (girls) for her friend so as to adjoin the unintended female with the man behind the trap is another system of sexual offenses in the area.
They further mentioned threatening girlfriends to have sexual intercourse is another problem of sexual offense by saying, "I will fi nd another girl if you didn't allow me", though she tries to convince him to wait until she completes her schooling. In line with this, men try to provoke girlfriends to make sex by saying "let me check whether you are a virgin or not". Literally, they are using the local language "Anoo durbummaakee hinballeessu nan borrosha malee" meaning "I don't defl ower you, but to check it." (31 years male governmental employee).
In the same notion, the perpetrators used physical force to have sexual intercourse with young females. This was complemented by the results from FGDs and one young woman

Discussions
This study used a community-based cross-sectional design using a mixed method of data collection method and highlighted the prevalence of sexual coercion at fi rst sexual debut and associated factors among young females in rural Ethiopia.
From the sexually experienced young females who reported the reasons for sexual commencement, sexual coercion is the highest and constituted 36.5%. This fi nding is consistent with the reports from different studies around the African countries indicated the prevalence of nonconsensual fi rst sexual experiences among young female of 33.1% in Uganda [22] and 27.5% in Nigeria [15]. Similar reports were shown in a study among young females in University and College students of Ethiopia that showed 43.3% in Bishoftu town [23], and 45.4% in Wolayita Sodo [18]. The slight variation in prevalence is attributed to the differences in the study settings in which the majority of the studies were conducted in school and urban settings. The fi nding is supported by the majority of FGD discussants approved for sexual coercion is a common reason for fi rst sexual debut for young females and further mentioned the issues considering as a serious problem of the study area.
In this study, the greater proportion (23.4%) of sexually coerced young females at fi rst sexual debut reported to have sex by the use of physical force/rape from the men. This is more than 12.7% in Addis Ababa [24]. This might be due to urbanrural disparities as rape is more common in rural communities were coerced by an unknown person to the victim (strangers).
These are comparable with similar reports from Kenya, which indicated that acquaintances and strangers perpetrated 21.8% and 8.2% of sexual coercion at fi rst sexual debut, respectively [25].
With regard to the estimated age of the perpetrators, only 16.8% of the victims were reported about the same age, while the majority reported for the age of the perpetrators was more than the victim's age, even 5.9% of them reported to about 10 years and above differences. This shows that some young females experience non-consensual sexual relationships with older, more powerful partners with whom they may feel unable to negotiate and perform unsafe sexual practices.
Married young women were less likely than single to report sexual coercion at fi rst sexual debut. This is supporting the fi ndings from similar studies in Kenya, which showed young females who had never been married, and those who did not live with their parents had a signifi cantly elevated risk of sexual coercion at fi rst sexual debut [25]. This is corroborating the notion that marriage is a safe refuge, especially for young females [25].
Young females living alone were signifi cantly at higher risk of getting sexual coercion at fi rst sexual debut than those living with their parents. Similar fi ndings were reported in other similar studies [14,24,26]. These were complemented by most discussants from FGDs mentioning for young females living away from parents are more vulnerable to sexual coercion. The study also showed a signifi cant association between earlier onset of sexual relations among young females and a greater risk of sexual coercion at fi rst sexual debut. This is supported by the reports from studies conducted elsewhere, explaining for being earlier age is vulnerable to unwanted sexual intercourse and young female who initiated sex at very young ages may also have experienced some sort of pressure either physical or verbal to have sex against their will [4,6,22,25].
In the study as the age of young people is increasing the probability of having sexual coercion at fi rst sexual debut is decreasing. Accordingly, those age group from 15-19 years were less likely to report sexual coercion at fi rst sexual debut compared to those aged 10-14 years. This is corroborating the results from different studies ( 27,28 ). This is attributed for young females at earlier age might lack a negotiating and convincing power to save themselves from being coerced.
This study also highlights the prominent role of drinking alcoholic beverages and chewing khat by young females that increases the risk of sexual coercion at fi rst sexual debut. These are evidenced by various studies on the subject area [4,17,23]. The plausible explanation for these might be behavioral risk factors that put young females at increased vulnerability to sexual coercion at fi rst sexual debut as usually people might not control themselves over sexual feelings by interpreting and effectively acting on warning signs. Moreover, all FGDs groups briefl y mentioned the role of alcohol and Khat use in association with sexual coercion at fi rst sexual debut for the young female in the study area.
As to the limitation of this study, due to its cross-sectional nature, it might not be possible to ascertain the direction of the cause-effect relationships between the study variables. Also, social desirability bias was inevitable due to the sensitive nature of the study. Despite these limitations, it is possible to generalize the fi ndings of this study to similar rural settings having similar socio-demographic characteristics in the country.

Authors' contribution
SG initiated and conducted the study. Analysis, write up and drafting of the manuscript is also done by SG.

Availability of data and materials
All data generated or analysed during this study are included.