Systematic Literature Review: Home

"A systematic review attempts to identify, appraise, and synthesize all the empirical evidence that meets pre-specified eligibility criteria to answer a research question. Researchers conducting systematic reviews use explicit methods aimed at minimizing bias to produce more reliable findings that can be used to inform decision-making." (Cochrane). We are pleased to welcome the Imam Abdulrahman Bin Faisal University Library's systematic review, LibGuide."

There are many types of systematic reviews, and they each have a slightly different purpose and methodology.

• Rapid review: narrow, quick search and assessment of a particular question
• Scoping review: evaluation of the potential size and scope of available literature
• Integrative review: includes qualitative, quantitative, and theoretical
• Umbrella review: a review of systematic reviews

For details of other review types, see

Grant M., Booth A. A typology of reviews: an analysis of 14 review types and associated methodologies (link opens in a new window). Health Information & Libraries Journal. 2009;26(2):91-108.

A systematic review is carried out in several stages. The timeline that follows shows the various phases.
When building the search and screening, it will save a ton of time if you take the time at the beginning of your review to ensure you are asking the right question and phrasing it correctly.
Every step of the review process should be documented, including search tactics, inclusion and exclusion criteria for screening, the process for evaluating publications, and the methods for extracting and critically analyzing data. This documentation will be useful when it comes time to write and submit the systematic review for publication or evaluation.

Dear Students and Faculty Members:

السلام عليكم ورحمة الله

I am pleased to share my practical experience of doing systematic reviews and meta-analyses with you.

Systematic reviews and meta-analyses are rigorous and detailed processes that require careful planning and execution. Here's an enhanced and more detailed guide based on your initial steps, including additional points for clarity and improvement:

1. Build a review team and timeline: A systematic review (SR) is a collaborative effort involving a team of three or more individuals with the necessary expertise, time, and ability to work together effectively. This team usually includes subject experts, search experts or librarians, and, in the case of a meta-analysis, a statistician or biostatistician. Before initiating the project, the team must become familiar with relevant software tools, such as citation management systems, screening platforms, and statistical applications. Depending on the complexity of the topic and the level of commitment from team members, the entire process of conducting a systematic review can take approximately six months or even longer. 

2. Define Your Research Question: A research question should be formulated using a structured framework such as PICO (Population, Intervention, Comparison, Outcome) or PICOS (including Study Design). The question must be specific, focused, and relevant to the field. A well-developed research question minimizes bias and provides clarity for readers, enabling them to clearly understand the scope and direction of the systematic review or meta-analysis.

3. Develop a protocol or plan: A systematic review should develop a detailed protocol that clearly outlines the review’s objectives, inclusion and exclusion criteria, data sources, and methodology to be applied. To ensure transparency and accountability and to reduce the risk of duplication, it is recommended that the protocol be registered in recognized databases such as PROSPERO.

4. Establish inclusion and exclusion criteria: When conducting a systematic review, it is essential to establish clear criteria for selecting studies. These criteria should be based on population characteristics, type of intervention, outcomes of interest, study design, publication timeline, and publication language. Developing well-defined inclusion and exclusion criteria helps to minimize bias while ensuring that the selected studies remain relevant and aligned with the research question.
 

5. Conduct a comprehensive literature search: A comprehensive search strategy is essential for a systematic review. This involves searching multiple databases such as PubMed, Medline, Web of Science, Scopus, Cochrane Library, Embase, Dimensions, and OpenAlex to identify all relevant studies. To ensure broad and accurate coverage, researchers should use a combination of search terms, keywords, and Boolean operators. It is equally important to document the entire search process, including the databases searched, the dates of each search, and the specific search terms used, to maintain transparency and reproducibility.
 

6. Manage and screen the studies (title/abstract): References should be managed using tools such as Silvi AI, Zotero, or Rayyan, which help organize citations and remove duplicates efficiently. The screening process should be carried out in two stages: Title and abstract screening, where studies that clearly fail to meet the inclusion criteria are excluded, followed by full-text screening, where complete articles are reviewed to confirm eligibility. Multiple reviewers should independently conduct the screening to enhance reliability and minimize bias, with any conflicts resolved through discussion or consensus.

7. Conduct Snowballing (Forward/Backward Citations): Snowballing is a valuable strategy for identifying additional relevant studies during a systematic review. It can be conducted in two ways. Manual snowballing involves reviewing relevant papers' reference lists from databases such as PubMed, Scopus, Web of Science, Google Scholar, and Semantic Scholar to identify cited works. Automatic snowballing utilizes digital tools, such as Citation Chaser, an R package, and a Shiny app that enables forward and backward citation tracking, as well as Inciteful. This interactive academic search engine uses citation networks to help researchers explore related literature more efficiently.
 

8. Conduct full-text Retrieval and screening: In the next stage of the systematic review process, the full texts of articles that have passed the title and abstract screening should be retrieved. Each article must be thoroughly reviewed to confirm eligibility based on the established inclusion and exclusion criteria. It is also essential to document the reasons for excluding articles at this stage to maintain transparency and ensure methodological rigor.
 

9. Extract data from selected studies: Data should be extracted systematically from all included studies, ideally using a standardized data extraction form to ensure consistency and accuracy. For qualitative synthesis, findings can be summarized narratively to highlight themes, patterns, and key insights. In cases where a meta-analysis is planned, only studies that provide sufficient statistical information should be selected to allow for quantitative pooling and meaningful comparisons.

10. Assess the quality of studies: To ensure the reliability of findings, it is essential to evaluate the methodological quality of each included study using appropriate tools. For randomized controlled trials (RCTs), the Cochrane Risk of Bias Tool is commonly applied, while for observational studies, the Newcastle-Ottawa Scale is recommended. The assessment should be thoroughly documented, and the results reported to highlight potential sources of bias, as well as the strengths and limitations of the evidence base.
 

11. Decide on a data synthesis approach: In a systematic review, evidence can be synthesized through two primary methods. Qualitative synthesis is applied when dealing with non-numerical data or studies that employ different designs, allowing findings to be summarized narratively. On the other hand, quantitative synthesis (meta-analysis) uses statistical methods to pool results from comparable studies to calculate an overall effect size. To ensure robustness, it is essential to assess heterogeneity among studies—commonly using the I² statistic—which helps guide the interpretation and reliability of the synthesis.
 

12. Assess the overall quality of evidence: To assess the quality of evidence in a systematic review, frameworks such as GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) should be used. This approach considers essential factors, including study limitations, findings consistency, results precision, and potential publication bias. It is also crucial to carefully determine what data must be extracted from the included studies to address the research questions effectively. Before proceeding, a pilot test of a draft data extraction form is recommended to ensure clarity, consistency, and reliability in the data collection process.
 

13. Conduct the Meta-Analysis (If applicable):

14. Interpret and discuss findings: In the final stage of a systematic review, the results should be summarized concerning the research question, providing clear answers supported by the evidence gathered. It is also essential to discuss the limitations and potential biases of the included studies and the review process itself. Finally, the review should highlight the implications for practice, policy, and future research, ensuring that the findings contribute meaningfully to the field and guide further investigation.

14. Report and publish the review:

15. Publish and disseminate findings

16. Contact and Suggestions: For suggestions, collaboration, or further discussion, I can be reached at shahmad@iau.edu.sa. This guide presents a refined roadmap for conducting systematic reviews and meta-analyses, emphasizing rigor, transparency, and reproducibility across all stages of the research process. The Cochrane Handbook for Systematic Reviews of Interventions serves as a fundamental reference for more in-depth support. At the same time, the PRISMA website provides essential reporting tools and templates to ensure the production of high-quality, methodologically sound systematic reviews.

Regards,
Dr. Shakil Ahmad