How to search practice guidelines efficiently: systematic review




Poster session 2


Monday 24 October 2016 - 15:30 to 16:00


All authors in correct order:

Chang X1, Wang C2, Luo X3, Wan M3, Li L4, Wang Z2, Wei D1, Chen Y1
1 Evidence-Based Medicine Center, School of Basic Medical Sciences, Lanzhou University, China
2 School of Public Health, Lanzhou University, China
3 School of Basic Medical Sciences, Lanzhou University, China
4 First Clinical Medical College of Lanzhou University, China
Presenting author and contact person

Presenting author:

Zixia Wang

Contact person:

Abstract text
Background: Guidelines are defined as systematically developed statements to assist practitioner and patient decisions about appropriate health care for specific clinical circumstances. Developing guidelines is expensive in terms of human resource and money. Therefore, if we complete the dissemination, implementation and application of guidelines, it will achieve the value of guidelines. However, the retrieval of guidelines is very important for their dissemination and implementation.
The AGREE (Appraisal of Guidelines for Research and Evaluation) enterprise recommends we search for guidelines using seven international guideline databases including NGC, NICE, SIGN, GIN, Canadian Medical Association Infobase, National Health and Medical Research Council (NHMRC), and eGuidelines. However, there are critical eligibility criteria for including guidelines in those databases. Therefore, if we only search the guideline databases, we will miss some guidelines. Then, the AGREE Enterprise still recommends searching PubMed in order to avoid omitting guidelines. However, there is no standard search strategy in PubMed for guidelines.

Objectives: To investigate the search strategy from reviews of guidelines to summarize a new search strategy.

Methods: We searched PubMed for reviews of guidelines. The search strategy is 'Guidelines as Topic' [MeSH] Filters: Meta-Analysis; Review; Systematic Reviews. Two reviewers screened the reviews of guidelines and abstracted data independently using a standard form. Disagreements were solved by discussion or the third reviewer.

Results: A total of 37,336 records were retrieved from PubMed. After removing 59 duplicates, 36,656 were excluded on the basis of the title or the abstract; full-text was necessary for evaluation of the remaining 621 articles. Finally, 250 articles were included. We are abstracting the data, and the final results will be published later.

Conclusions: An appropriate search strategy for guidelines will be helpful for dissemination and implementation of guidelines, and benefit the guideline developers, guideline targeting audiences, methodologists focused on guidelines, etc.