Review protocol development
The Preferred Reporting Items for Systematic review and Meta-analyses (PRISMA) guideline will be employed to develop the review protocol [28] and PRISMA-P 2015 checklist format will be used to report the review procedure [29] (Additional file 1).
PECO search guide
Population: Young people (10–24 years old) [30].
Exposure: exposures are determinants that increase or decrease the likelihood of preventive practice toward sexually transmitted infections.
Comparison: The reference group for each determinant in each study will be the comparison variable. It may include good knowledge versus poor, positive attitude versus negative, education versus no education, access to information versus no accesses, consistent use of condoms versus not, etc.
Outcome: The primary outcome variable is the overall prevalence of good preventive practices of sexually transmitted infections (STIs) among young peoples in Ethiopia. The secondary outcome variables is the determinants of preventive practices of sexually transmitted infections (STIs) among young people in Ethiopia. Preventive practices of sexually transmitted infection refer the use of at least in one of the following activities; condom use, faithfulness, absenteeism from sex, decrease sexual partners, avoid sex with prostitute etc.
Data source and searching strategies
Databases such as PubMed, CINAH, Google, and Google Scholar will be used to search published and unpublished studies from May 15-12 August/2021. The two authors (EW and SB) will retrieve the studies. In addition, a cross-reference search will be employed to add other related studies from the final included studies that may miss in the databases search. The search term or keywords will be used using Boolean operator and the search string will be adapted based on the specific requirement of the database to identify relevant studies. (Additional file 2). The retrieve studies will be exported to Endnote version 8 reference manager software [31].
Eligibility criteria
All observational studies (cross-sectional, case-control, and cohort) will be included in the systematic review and meta-analysis study. Studies that reported the prevalence of preventive practices of sexually transmitted infection (STIs) and its determinants among young people in Ethiopia will be included. Moreover, we will also include studies that reported only the overall prevalence of good preventive practices of sexually transmitted infections (STIs) or at least one of the component of preventive practices of sexually transmitted infections. Both institutional and community-based studies on the prevalence of preventive practices of STIs will be included. Studies that only address the qualitative approach will be excluded from the study. However, if studies examine both quantitative and qualitative study findings, only the quantitative results will be considered. Studies published other than the English language, expert opinions, conferences, national surveys, and case reports will be excluded from the study. Restrictions to the date of publication will not be made.
Selection of studies
The two authors (EW and SB) will independently screen the studies based on the titles and abstract. Duplicates, irrelevant titles, and abstracts studies will be removed from the citation manager. Further quality evaluation will be made for full-text studies, and studies that meet the criteria of quality measurement standards will be included in systematic review and meta-analysis. The discussion will be made with the third author (MA) to solve any disagreement among the two reviewers (EW and SB) during the review process. The selection process will be presented using a PRISMA flow diagram (Additional file 3).
Quality assessment
The Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument (JBI-MAStARI) will be employed to assess the quality and validity of the study [32]. The quality assessment will focus on clear inclusion criteria, study subjects and setting, standard measurement criteria, exposure and outcomes measurement, and appropriate statistical analysis (Additional 4). The quality of the studies will be assessed independently by the two authors (EW and SB). Studies 50% and above of the quality scale score will be considered for the final systematic review and meta-analysis study. Differences among reviewers during the quality review will be further discussed by the third author (MA).
Data extraction
Data extraction template form on Microsoft Excel (2016) will be prepared. Before the beginning of the actual data extraction, the data extraction pilot will be done in the data extraction template form. The first author's name, publication year, the study area, study design, sample size, associated factors, odds ratio, and prevalence of the studies will be extracted from the studies. In addition, the logarithm and standard error (SE) of the prevalence and odds ratio will be calculated on Microsoft Excel template form. The two authors (EW and SB) will extract the data independently. The discussion will be mad for any difference with a third author (MA) to reach on consensus. Communication will be made with the corresponding author for studies, in case of missing data or incomplete reports.
Data synthesis and statistical analysis
A narrative synthesis of results will be made for the included studies. A summary table and graph will be presented to describe the characteristics of the included studies results. A random-effects model [33] will be employed to estimate the overall pooled prevalence of preventive practices of sexually transmitted infections (STIs) and identify its determinants among youths in Ethiopia. A 95% CI will be used to declare the statistical significance. Statistical heterogeneity will be checked using the Cochran Q test [34] and I2 statistics [35]. I2 values represent 25% low, 50% moderate, and 75% substantial heterogeneity. Subgroup analyses and meta-regression based on regions will be done to identify the sources of heterogeneity among the studies at p-value <0.05. Sensitivity analysis will be done to assess the effect of studies on the overall effect estimate. The presence of publication bias will be checked using funnel plot observation[36], Egger’s, and Beggar’s statistical tests [37].