Task |
Process |
Category |
|---|---|---|
| 1. Formulate review question | Decide on the research question for the review. | Preparation |
| 2. Locate prior Systematic Reviews on topic | Has an SR already been done on the research question? Look in databases like PubMed or in PROSPERO to see if there are existing systematic reviews that answer your research question. | Preparation |
| 3. Write and register the protocol | Write up an objective, that is reproduceable, with a methodology. Register the research topic in PROSPERO. Provide an objective, methodology and the inclusion/exclusion criteria. | Writing |
| 4. Create preliminary search strategy | What databases will be searched? What are your search concepts? Develop a list of the search concepts with keywords (along with their synonyms) and controlled vocabulary (example: MeSH) for each database when appropriate. | Preparation |
| 5. Run the searches | Take the search strategy and apply it across all the databases that will be searched. Locate all citations, including potential irrelevant ones. | Retrieve |
| 6. Remove duplicate articles (Deduplicate results) | Remove identical records retrieved from multiple databases. | Retrieve |
| 7. Screen Title/Abstracts | Based on inclusion/exclusion criteria, remove irrelevant articles based on title and abstract. | Appraise |
| 8. Pull full-text articles | Obtain full-text PDFs of articles that meet the inclusion criteria. Place interlibrary loan requests as needed. | Retrieve |
| 9. Screen full-text | Based on inclusion/exclusion criteria, remove irrelevant articles based on reading the article. | Appraise |
| 10. Review references (Snowball searching) | Pull bibliographies from the included articles. According to Cochrane, searching reference lists is mandatory; reference lists in included studies and any identified relevant systematic reviews should be checked. | Appraise |
| 11. Synthesize/Extract data | Convert extracted data to common representation. | Synthesis |
| 12. Assess bias | Determine bias, limitations and flaws in the design, conduct, or analysis of each included study. | Appraisal |
| 13. Repeat search strategy. | Run the searches a second time to find latest literature since the initial search. This is required if the first search was done over a year ago. | Retrieval |
| 14. Meta-analysis | If conducting a meta-analysis, statistically combine results from all studies. | Synthesis |
| 15. Write | Write the manuscript. Submit to journals if seeking publication. | Writing |
Adapted from: Tsafnat, G., Glasziou, P., Choong, M. K., Dunn, A., Galgani, F., & Coiera, E. (2014). Systematic review automation technologies. Systematic Reviews, 3(1), 74–74. https://doi.org/10.1186/2046-4053-3-74