Premarital sexual practice and its associated factors in Ethiopia. A systematic review and meta-analysis

Background Globally, large numbers of adolescents engage in premarital sexual practice. Youths who begin early sexual activity are more likely to be exposed for high-risk of having multiple sexual partners, infection with Human Immune Deficiency Virus (HIV) and other sexual transmitted diseases. There are pocket studies which reported meager and fragmented evidences across different parts of Ethiopia related with this issue, hence the aim of this review was also to estimate those meager evidences and come up with concrete information at the national level. Methods articles were retrived through search engines such as: PubMed/MEDLINE, EMBASE, CINAHL, Google Scholar, HINARI portal, and Cochrane Library using the preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) guidelines. Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument (JBI-MAStARI) was used for critical appraisal of the included articles. Random effect model analysis technique was considered to estimate the pooled prevalence of premarital sexual practice with their respective odds ratio (OR) and 95% confidence interval (CI). Cochran’s Q statistic, Egger’s and Begg’s test and meta regression were carried out to assess heterogeneity, publication bias and to identify associated factors respectively. Results Exactly 24 articles with total sample size 14,872 individuals were included for this review. The pooled estimate of premarital sexual practice at the national level was 30.98% (95% CI: 26.74, 35.21). Being female sex (OR=1.55: 95% CI; 1.01, 2.37), age >18yrs (OR=2.39: 95% CI; 1.29, 4.42), being alcohol abuser with (OR=3.47; 95% CI; 1.94, 6.19) and exposed to pornography (OR=3.24:95% CI; 1.95, 5.38) were the associated factors of premarital sexual practice in Ethiopia. Conclusion The pooled prevalence of premarital sexual practice in Ethiopia was high in comparison So that it is the strong recommendations to provide an integrated youth’s reproductive and sexual health services on the local context to decrease their exposure for premarital sexual practices and its negative impact on future life opportunities.

comparison with other countries. So that it is the strong recommendations of the authors' to provide an integrated youth's reproductive and sexual health services on the local context to decrease their exposure for premarital sexual practices and its negative impact on future life opportunities. Keywords: Pooled; Premarital; Sexual; Practice; Prevalence; Ethiopia .

Background
Marriage and sexual activity help to determine the extent to which women are exposed to the risk of pregnancy. However, the timing and circumstances of marriage and sexual activity also have profound consequences for women's and men's lives [1].
Globally, large number of adolescents engage in premarital sexual practice[2, 3]. As they are so young and less likely to use condom, unprotected or unsafe sexual practice is high [4,5]. Among youths who practiced premarital sexual practice, only very few numbers were continued as husband later on, while majority had sex with love even end up with prostitute [6]. Youths who begin early sexual activity are more likely to be exposed for high-risk of having multiple sexual partners [3,7].
Premarital sexual practice among youth has a number of adverse effects and consequences interlinked with sexual and reproductive problems. Of which, early sexual debut increases young peoples' risk for infection with HIV and other STIs [7,8], greater risk of unwanted pregnancy and prone for termination of the pregnancy through induced abortion [9]. All those complicated situations in turn increases the burden of maternal mortality and morbidity [10,11]. Moreover, it is also causes regrets, loss of self-respect, loss of family support, depression and victims to rituals among others [12].
In Ethiopia, the median age at first sexual intercourse is 0.5 years earlier than the median age at first marriage, which also indicates that both women and men engage in sex before marriage. Conversely, this increases the incidence of morbidity and mortality for both the mother and the child as a result of teenage pregnancy [1,13,14]. Giving birth during adolescence is known to have adverse social consequences in particular of poor educational attainment and early drop out of school [1,8,12,15]. Involvement in gang activities, drinking, smoking, living in low cost housing, dropping out of school, previously sexually abuse, peer pressure and having friends were explored as factors of premarital sexual practice [10, 16, 17].
In Ethiopia different pocket studies assessed the premarital sexual practice and associated factors [5,9,11,. However, the reports from those pocket studies on premarital sexual practice were represented with inconclusive and inconsistent findings.
Hence the aim of this review was also to estimate those meager evidences and come up with concrete information at the national level.

Searching strategies
To identify articles, a comprehensive search of PubMed/MEDLINE, EMBASE, CINAHL, Google Scholar, HINARI portal (which includes the SCOPUS, African Index Medicus, and African Journals Online databases), and Cochrane Library was carried out. The following searching terms were applied for PubMed advanced search: prevalence, practice, premarital sex, associated factors, risk factors, predictors and Ethiopia using OR, AND bullets. In addition, studies found on local shelves and institutional repositories were considered. The reference lists of already identified studies were screened to retrieve additional articles.
The review was reported using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) checklist guidelines [40] (additional file-1). Articles conducted till July 30, 2019 on premarital sexual practice in Ethiopia were included in this meta-analysis. Endnote citation manager software version X7 for Windows was utilized to collect and organize search outcomes and for removal of duplicate articles.

Inclusion criteria
Population: studies conducted on unmarried women were included.
Study area: the review considered only studies conducted in Ethiopia.
Language: articles reported in English language were eligible for this systematic review and meta-analysis. All available studies were included without restricting to a specific study design. In addition, predictors were taken as an exposure and premarital sexual practice as final outcome.

Exclusion criteria
After careful reviewing of the searched articles, irrelevant studies and those didn't report outcome of interest were excluded from this systematic review and meta-analysis. In addition, during the article selection process, studies, which were not fully accessible (full text available), were excluded. However, before excluding the articles, we attempted to contact the primary author at least two times through email.

Outcome of interest
The outcome of this systematic review and meta-analysis was to estimate the pooled prevalence of premarital sexual practice in Ethiopia. In addition, we identified factors associated with premarital sexual practice. From those studies factors such as; sex of the participant (male v s female), resident (urban v s rural), age (<18yrs v s ≥18yrs), pornography film (yes v s no), alcohol (yes v s no) and living alone or with family were considered to estimate the association factors.

Data extraction
Two authors (TG and AN) extract the data independently using the excel spreadsheet extraction tool. Authors, Region, study setting, study year, study design, sample size, response rate, participant's mean age, and prevalence for the primary outcome of interest and common associated factors extracted using two by two tables and logs odds ratio for each factors for our secondary outcome of interest were computed in the extraction tool.
Discrepancies between authors were discussed to reach consensus. For final analysis the authors considered articles which fulfilled the already settled criteria.

Quality assessment
The database search results were exported and duplicate articles were removed using EndNote software (version X7; Thomson Reuters, New York, NY). We used the Newcastle-Ottawa quality assessment scale adapted for cross-sectional studies [41,42]. The scale is used to score the articles under three categories: Selection (score 0-5); Comparability (score 0-2); and Outcome (score 0-3); total score range 0-10.
The first section scored focuses on the methodological quality of each study (i.e., sample size, response rate, and sampling technique). The second section of the tool considers the comparability of different outcome groups in the study based on the study design and analysis. The last section deal about the outcomes and statistical analysis of the original study. Two authors (TG and AN) gave score for each primary studies. The third author was considered in case of disagreement. Then studies with a score of ≥6 out of 10 were considered as high quality after reviewing different relevant literatures. Identified articles with methodological problems and incomplete reporting of results in the full text were excluded from this systematic review and meta-analysis (additional file 2).

Data analysis
Information computed from retrieved articles was extracted using Microsoft Excel spreadsheet form and exported to STATA (version 14; Stata Corp, College Station, TX) for further analysis. Cochran's Q statistic was quantified using inverse variance (I 2 ) with their corresponding p -value. Heterogeneity was considered when p-value less than 0.05. A value of 25%, 50%, and 75% were used to declare the heterogeneity test as low, medium and high heterogeneity [43]. In addition, publication bias was checked using Egger's and Begg's tests [44,45] and a p-value less than 0.05 were considered as statistically significance. The estimated pooled prevalence of premarital sexual practice was computed using forest plots with the 95% CI. Odds ratio was considered as measure of association to identify the factors associated with premarital sexual practice in Ethiopia.

Results
A total of 230 articles were searched using electronic data base of which 50 articles were removed due to duplication. From the remaining articles, 151 studies were excluded because of irrelevancy. Finally 5 articles [36,[46][47][48][49] were removed as a reason of inaccessible of its full documents. Finally, 24 articles were included for the final systematic review and meta-analysis (figure-1).  and one articles [31] conducted in Addis Ababa city administration. Except one study which was conducted at community level in Tigray region [23], all other studies were done at the institutional level. After critical appraisal of each articles based on Newcastle-Ottawa quality assessment scale, both in peer and independently, each studies scored in the range of 6-up to 9 values.

Sub group analysis
Since, this review is exhibited with substantial heterogeneity, subgroup analysis based on study region, study year, participants mean age and sample size were considered to identify possible source of heterogeneity across studies (table-2). However the subgroup analysis indicated that the possible source of heterogeneity was not due to study region, study year, participants mean age and sample size they used. In addition publication bias was assed using Egger's and Begg's tests and the value significant at p-value of 0.021 and 0.004 respectively. So that the trim fill meta-analysis [50]was done to account for the publication bias. Based on this analysis, the prevalence of premarital sexual practice was 30.98% (95% CI: 26.74, 35.21) and no significant change was exhibited as compared with the main meta-analysis.

Meta regression
In addition to subgroup analysis and publication bias, Meta regression was also assumed by considering both continuous and categorical data to identify associated factors of heterogeneity for the pooled prevalence of premarital sexual practice.  -3 and figure-4).  Although premarital sex is socially unacceptable in Ethiopia [1], the review has shown that the proportion of individuals having sexual intercourse before marriage is considerably high. Due to the sensitive nature of the issue [51], this proportion may still be underreported. The review revealed that nearly one third of youths practiced premarital sex in Ethiopia. The new generation seems more open towards premarital sexual activities.
Young age is more liberal towards their sexual behaviors [17].

Adolescent finds the way to break conventional obstacles and looks for the information by themselves on the issues which have significant influence in their life as well on their
sexuality [52]. The prevalence of premarital sexual practice in this systematic review and Even though cultural beliefs discourage premarital sexual practices[27], Ethiopian women exposed to risk of sexual practice and pregnancy earlier [1]. This systematic review and meta-analysis also indicated that exposing to pornography, alcohol drinking, being female and age more than 18 years were found to be significantly associated with premarital sexual practice in Ethiopia. Individuals who had exposed pornography film were more likely to have premarital sexual practice than those didn't exposed pornography. They may watch pornographic films to fulfill their desires to learn about sexual issues and may get opportunity like out of family supervision and stimulate sexual desires. Consistent findings were also reported from Asian countries and Vietnam[6, 10]. Late age youths (age more than 18yrs) were more likely to experience premarital sexual practice than early age youths (age less than 18yrs). In line with this systematic review and meta-analysis, a report from Nepal [54] also reported that youths of age 20-24yrs were more likely to expose for premarital sexual practice than youths age from 18-19yrs.
Moreover, the likelihood of premarital sexual practice among female was higher, when compared with male counterparts. This is because the effect of culture forces females to be exposed towards premarital sex than their male counterparts and most girls have a belief that they become ready to do anything to make their love succeed [17]. In contrast to the report from this systematic review and meta-analysis, studies from Nepal and Vietnam showed male were more likely to be involved in premarital sex than female[4, 6, 54], because mostly men have freedom to decide on major tasks in their life.
Once more, similar with the reports from Asian countries and Nepal[10, 54], Youths who drink alcohol were more likely get involved in premarital sexual practice than those who didn't drink alcohol. This can be possibly explained by that risky sexual behavior such as smoking and alcohol drinking is associated with higher approval of premarital sexual practice [5, 6, 17].

Conclusion
This systematic review and meta-analysis indicated a significant number of (