Evidence Summaries are based upon a systematic search for the highest level and quality of evidence. Systematic means searching a predetermined suite of databases and the use of search terms that are calibrated to the database and topic as described below. Contributors conduct searches for evidence that informs best practices relevant to the topic.
The following electronic databases MUST be searched for source information using a range of keywords and subject headings appropriate to the specific topic:
- JBI Evidence Synthesis
- Cochrane Library
- Medline - searched via PubMed or another platform such as Ovid, EBSCO, etc.
- CINAHL (Cumulative Index to Nursing and Allied Health Literature)
- Additional databases may be searched, where relevant, for specific topics e.g. PsycINFO (mental health), Physiotherapy Evidence Database (PEDro), etc.
For resources to improve your literature searching, please see this guide to searching health databases. This PubMed tutorial on conducting a search may also be helpful, for example learning how to use Boolean operators.
Whether developing a new evidence summary or updating an existing evidence summary, a three-phase search strategy must be undertaken:
- Conduct a search to establish if any systematic reviews or evidence-based guidelines (see searching for guidelines below) have been published on the topic. If additional reviews or guidelines have been published, these are incorporated into the Evidence Summary as per the guidance below; no further searching is required if the included systematic reviews and evidence-based guidelines answer the clinical question.
- If the first search establishes that no new systematic reviews have been published on the topic, a broader search is undertaken for primary research. The aim of this search is to retrieve the highest level of relevant evidence to answer the clinical question. If multiple new primary research studies are found, incorporate the most relevant studies into the evidence summary, assessing how relevant the information is, the level of the new evidence (see JBI Levels of Evidence), and the quality and size of the study.
- If no research studies on the topic are found, then a search for expert opinion evidence is conducted.
Searching for evidence-based guidelines and expert opinion papers
In the absence of high-quality systematic reviews and other, higher-level evidence it may be necessary to include expert opinion papers. Note that only published expertise is accepted, personal communications are not considered appropriate for inclusion in a JBI evidence summary, when reporting expert opinion, the level of evidence should be clearly identified, and full details of the citation provided in the reference list.
When searching for evidence-based guidelines or expert opinion papers, the above databases should be searched first; however, it may be necessary to conduct a wider internet search for reputable organisations/affiliations. The following list is a guide to potentially relevant sites that you may search for evidence-based guidelines or expert opinion. This is not a comprehensive list and it is necessary to tailor the search to the specific topic area.
- World Health Organization (WHO)
- National Institute of Clinical Excellence (NICE)
- The Scottish Intercollegiate Guidelines Network (SIGN)
- National Health and Medical Research Council (NHMRC)
- Relevant organisation and associations related to the topic area (e.g. European Society of Cardiology for cardiac-related topics)
- It may be necessary to conduct a general search using Google.
Please note that evidence from books/textbooks or spoken/email communication is not accepted. Other content-related decisions are up to the discretion of the contributor, but the evidence reported must be related to answering the clinical question.
NOTE FOR UPDATING RESOURCES
- While the search should focus on evidence published since the Evidence Summary was last updated, do not discount relevant evidence from the previous five years.
- It is also important to establish if any systematic reviews or guidelines already included in the Evidence Summary have been updated; if so, the reference should be replaced with the updated version.
The date range, databases searched, and search terms used and are reported in the Technical Development Report.