How many network meta-analyses were adopted by clinical practice guidelines? A cross-sectional survey




Poster session 2


Monday 17 September 2018 - 12:30 to 14:00

All authors in correct order:

Bian Z1, Yao L1, Tian R1, Yang Z1
1 Chinese Medicine Faculty of Hong Kong Baptist University, Kowloon Tong, Hong Kong, China
Presenting author and contact person

Presenting author:

Liang Yao

Contact person:

Abstract text
The development of clinical practice guidelines should be based on systematic reviews. Network meta-analyses are useful in clinical practice guidelines when there is no evidence directly comparing interventions of interest (indirect treatment comparison) or various interventions exist for one clinical question.

This study aims to analyse the proportion of network meta-analyses used in clinical practice guidelines.

We searched the PubMed database to obtain network meta-analyses, of which we randomly selected a sample using SPSS 19.0. We then collected the articles that cited the network meta-analysis sample using Google and Web of Science.

We retrieved a total of 289 network meta-analyses from PubMed and randomly sampled 100. We found that 2026 articles had cited these network meta-analyses, among which 50 (2%) were clinical practice guidelines; this included 12 NGC guidelines and 2 NICE guidelines. All of these clinical practice guidelines citing network meta-analyses were conducted or updated in 2014-2017.

Network meta-analyses are valuable when multiple pairwise comparisons are presented in clinical practice guidelines. However, we found that a low proportion of network meta-analyses were used in guidelines. Further research on how to improve the application of network meta-analyses in clinical practice guidelines would be valuable.

Patient or healthcare consumer involvement:

Relevance to patients and consumers: