Reporting
The Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guideline[34] used to report this meta-analysis(Additional file 1 research checklist).
Literature search
The databases for the search were Medline (PubMed), Web of Science, and Google Scholar. The terms for the search were pre-defined for a comprehensive search strategy. These included all fields within records and Medical Subject Headings (MeSH terms). In the Boolean operator, within each axis, we combined keywords with the “OR” operator. Then we linked the search strategies for the two axes with the “AND” operator. The search terms used for the search were “Preoperative anxiety” OR “anxiety” OR “surgical anxiety” OR “pre-surgical anxiety” OR “preoperative worry” OR “concerns” OR “nervousness” AND “prevalence” OR “magnitude” OR “burden” AND “Ethiopia”. The specific searching detail in PubMed with MeSH terms was (“Preoperative anxiety” [MeSH Terms] OR “anxiety”[MeSH Terms] OR “surgical anxiety”[MeSH Terms] OR “pre-surgical anxiety”[MeSH Terms] OR “preoperative worry”[MeSH Terms] OR “concerns”[MeSH Terms] OR “nervousness”[MeSH Terms] AND “prevalence” [All Fields]) OR “magnitude”[MeSH Terms]OR “burden”[MeSH Terms] AND (“Ethiopia” [MeSH Terms] by the date 02/03/2020.
The publication year of the studies was not limited during the search.
Study selection
All retrieved studies were exported to Endnote version 7 reference manager. It is the study selection method that we used to remove duplicated studies.
The retrieved articles were screened according to pre-defined inclusion and exclusion criteria. Discussion and/or involvement of the third reviewer resolved any disagreements.
Eligibility criteria
Inclusion criteria
Included studies were articles that reported the prevalence of preoperative anxiety. It also included studies published in English and studies conducted only in Ethiopia.
Exclusion criteria
Excluded criteria were articles without full-text available and qualitative studies. Other excluded criteria were any reviews, commentaries, consultants’ corners, letters, and conference abstracts.
Quality assessment
We used Joanna Brigg’s Institute (JBI) quality appraisal criteria[35]. It is the assessment tool used to check the quality of each article. The tool consists of nine major items. The first item is appropriate to the sample frame. The second is the appropriate sampling technique. The third is the adequacy of the sample size. The fourth is a description of the study subjects and settings. The fifth is enough coverage of data analysis. The sixth is the validity of the method for identification of the condition. The seventh item is a standard and reliable measurement for all participants. The eighth is the appropriateness of statistical analysis. And the last item is adequacy and management of response rate. Studies considered low-risk when it would fit 5 or above quality assessment checklists.
Data extraction
A standardized form used to extract data by two authors. The following information from each article was extracted. Such as first author, and publication year, the study design, and study population. The locations of the study were also extracted.
Outcome measurement
The major outcome is to determine the prevalence of preoperative anxiety in Ethiopia. It calculated as dividing the numbers of patients who develop anxiety to the total number of patients multiply by 100. A total number of patients refer to all adult elective patients during the study period. Preoperative anxiety is an event, a person presents with signs and symptoms of the anxiety[36]
Data analysis
The required data were collected using a Microsoft Excel 2010 workbook form. Then, the STATA Version11 software was used to analyze the data. The original articles presented using tables and forest plots. A weighted inverse variance random-effects model[37] used to estimate the pooled prevalence. I2 statistics used to assess the percentage of total variation across studies [38]. I2 ≤ 25% suggested more homogeneity. 25% < I2 ≤ 75% suggested moderate heterogeneity, and I2 > 75% suggested high heterogeneity[38]. Egger’s regression test was also used to assess publication bias [39]. Furthermore, the sub-group analysis carried out based on the region of studies. This reduces the random discrepancies between the point estimates of the primary study.