Premarital sexual intercourse and associated factors among adolescent students in Debre-Markos town secondary and preparatory schools, north west Ethiopia, 2017

Objective To assess the magnitude and factors associated with premarital sexual intercourse among adolescent students of the secondary and preparatory school in Debre-Markos town, northwest Ethiopia, 2017. Results Among secondary and preparatory school adolescent students, 31.3% reported pre-marital sexual intercourse. This shows that premarital sexual intercourse among secondary and preparatory school adolescents is high. Significantly associated factors were: being male (AOR = 1.9, 95% CI 1.21, 2.93), having pocket money (AOR = 3.1, 95% CI 2, 4.81), adolescents who did not discuss sexual issue with close friends (AOR = 8.6, 95% CI 5.27, 13.91) and peer pressure (AOR = 7.7, 95% CI 3.73, 15.69).


Introduction
Premarital sex is penetrative vaginal intercourse performed between couples before formal marriage [1,2]. World Health Organization (WHO) defines adolescent as persons between the age group of 10-19 years old [3]. Adolescence is the period of transmission from childhood to maturity and is characterized by spurt of physical, mental, emotional, social and psychosexual development [4].
Adolescents are a growing and larger segment of the population of developing countries and an estimated 1.2 billion young people in the world, 85% live in developing countries [5]. Nearly 85% of the world's adolescent population live in developing countries and in some sub-Saharan countries, population below 15 years of age is five times greater than the population over 55 years of age [6].
The adolescent years are the time of rapid growth, exploration, and risk-taking. In many countries, an average of 29% of boys and 23% of girls are sexually active including premarital sex [7].
Many adolescents face pressures to use alcohol, cigarettes, or other drugs and to initiate sexual relationships at earlier ages, to put themselves at high risk for intentional and unintentional injuries and risky sexual behaviors [8]. Females, particularly adolescent girls may end up with unwanted pregnancies, abortions, teenage deliveries, and various complications of these including death. Moreover, the girls may drop out from school to look after their children, and in most cases, they become economically reliant on upon their parents [1].
Nearly 70% of premature deaths among adults can be linked to behaviors that were initiated during adolescence [9]. Unwanted pregnancy can be associated with higher likelihood of early motherhood, unsafe abortion, and other pregnancy-related complications [10].
Several studies in sub-Saharan Africa have also documented high and increasing premarital sexual activities among adolescents [8]. According to EDHS 2016, 13 percent of women age 15-19 in Ethiopia have begun childbearing [11]. In Amhara region, pre-marital sexual debut was reported as early as 12 to 13 years [10].
Studies have documented that early sexual initiators were more likely to report undesired consequences of sexual initiation such as teenage motherhood, not using a condom at first sex and sexually transmitted infections (STIs). Adolescents are also likely to have an intimate partner who is five or more years older and be involved in multiple sexual partnerships [12].
Even though some studies were conducted on this topic, most of them focused on only females [5,10,13], youth [14,15] and university students [7,8], there is a gap of including males, married females and adolescents. Therefore, this study tried to fill the above gaps.

Study design and setting
An institutional based cross-sectional study design was conducted among secondary and preparatory school adolescent students in Debre-Markos town from November 23-27, 2017. The town is located at about 295 km to the capital city of Ethiopia. Based on the 2007 national census conducted by the Central Statistical Agency of Ethiopia (CSA), this town has a total population of 62,497, of whom 29,921 were men. The majority Ethiopian Orthodox Christianity followers, with 97.03% reporting [16]. There are three secondary and two preparatory schools in the town.

Sample size and sampling procedure
A single population proportion formula was used to calculate the sample size of 624 by taking the following assumptions. From the previous study conducted in Jimma town on premarital sexual practice among school adolescents [17], 25.27%, 95% CI, 5% marginal error and n = ((Zα/2) 2 p (1 − p))/W 2 . Adding a 15% non-response rate and design effect of 2, the total sample size required was 624.
Multi-stage stratified sampling technique used to select adolescent students. All regular adolescent students (secondary and preparatory) attending class at the time of the survey in Debre-Markos secondary and preparatory school divided into different strata. Grade considered as strata. The number of adolescent students from each grade level and sections according to their sex identified by using the name and sex list of each section.

Operational definitions
Age at initial sexual contact is age at first intercourse (vaginal-penile penetration).
Early sexual initiation was taken as an experience of first intercourse before 18 years of age.
Sexually active A student who had a penetrative sexual intercourse (vaginal) at least once prior to the study.
Peer pressure when the individual said yes/no to question saying "did your friend initiate you to do sex?" [18].
Pocket money when the individual said yes/no to question saying "did you have pocket money?" [19].

Data collection instrument and process
Data were collected using a semi-structured, pre-tested and self-administered questionnaire adapted from the literatures. The data collection tool was prepared in English and then translated into local language Amharic and finally returned to English by English language expertise. Four midwives were involved in the data collection process.
Appropriate information and instructions were given on the objective, the relevance of the study, confidentiality of information, respondent's rights, informed consent, and technique of data collection and 1-day training was given to data collectors and the supervisor on data collection. Before the actual data collection, pre-test was done on 5% of students in Dejen secondary and preparatory school which was out of the study setting but nearby. The collected data checked for completeness and clarity by principal investigator and supervisor. Privacy and confidentiality of the respondents were maintained throughout the data collection period.

Data analysis
Data coded and entered into a computer using Epi info version 7.2.0.1 and checked for completeness and transferred to SPSS version 20 for analysis. Descriptive statistics like frequencies, percentage, proportion and mean computed. Bi-variate logistic regression used to identify variables that crudely associated and variables with p-values less than or equal to 0.05 fitted to multiple logistic regression. Then association between dependent and independent variables was assessed using adjusted odds ratio (AOR), 95% CI and p value of ≤ 0.05 considered statistically significant.

Socio-demographic characteristics
From the selected 624 school adolescents, a total of 600, adolescents aged between 10 and 19 completed the questionnaire while 24 refused to participate in the study, giving a response rate of 96.15%. Three hundred four   (Table 1).

Parental characteristics
From the total participants, three-hundred thirty three (55.4%) responded that they were living with their mothers and fathers, 583 (97.2%) and 552 (92%) of respondents reported that their mothers and fathers were alive respectively. Two hundred sixty-four (44.3%) reported that they did not know their parents' income and 263 (44.1%) participants responded this as greater than 53.6 USA dollars ( Table 2).

Habits of the respondents
Greater than two third, 413 (68.8%) of the participants responded that they were not alcohol users but 62 (10.3%) reported that they were khat users.

Sexual behavior of respondents
Two hundred forty seven (41.2%) of the participants responded that they usually watch films and magazines having sexual contents and nearly one third, 188 (31.3%) of the whole participants responded that they had girl/ boyfriends and practiced sexual intercourse.

Reproductive health
From the total respondents, only 95 (15.8%) participants responded that they had sexual issue discussion with Table 3 Bivariable and multivariable analysis of factors associated with premarital d sexual intercourse among adolescent students in Debre-Markos secondary and preparatory school from Nov. 23-27, 2017 (n = 600) * p-value < 0.05, ** p-value < 0.01 and *** p-value < 0.001

Associated factors of premarital sexual a intercourse
Crudely associated variables were: sex, current residence, pocket money, with whom usually live, mother alive, father alive, place of parent's live, discussion the sexual issue with close friends and peer pressure. Independently and positively associated variables in adjusted analysis were: being male, having pocket money, adolescents who did not discuss the sexual issue with close friends and Peer pressure ( Table 3).
But this study's finding was lower than the study from Nepal, 36.5% [20]. This disparity could be justified by the difference in a background of the study participants and variation in the study areas.
One of the predictor variables in this study was sex. It shows that male students were more engaged for premarital sexual intercourse (AOR = 1.9, 95% CI 1.21, 2.9). It was consistent with studies from Yabello [15], Bahir Dar [22] and Malaysia [23]. It might be due to males have more freedom in sexual engagement than females.
The other variable that was positively associated with premarital sexual intercourse was having pocket money. It triggers adolescent student to have sexual intercourse (AOR = 3, 95% CI 1.9, 4.8). It was similar with the studies from Arba Minch [24] and Jimma [19]. It could be due to financially equipped students are likely to drink alcohol, to date their opposite friends and to buy the porn films that all can trigger for sexual intercourse.
And also, the other positive predicator variable for premarital sexual intercourse was having discussions with close friends. Those who did not have discussions with their close friends were more engaged with premarital sexual intercourse (AOR = 8.6, 95% CI 5.28, 13.9). It could be explained as those who have discussions are likely to be knowledgeable about premarital sexual intercourse risks including HIV transmission which can make adolescent students to abstain from sexual intercourse.
The additional positive predictor variable for premarital sexual intercourse was peer pressure. Those who were pressurized by their friends were more engaged with premarital sexual intercourse (AOR = 7.7, 95% CI 3.7, 15.7). It was consistent with studies from Bahir-Dar [10], Maichew [5], Alamata [21], Gondar and Metema [13]. It could be due to the fact that peers in adolescents is an important factor to influence personality and behavior changes.

Limitations
It was good if the data collection tools were triangulated with the qualitative data collection techniques like in-depth interview.

Authors' contributions
GKB brought the idea. GKB, KTA and GLA equally contributed on proposal development, data collection process, data management and analysis, and write up. All authors read and approved the final manuscript.