Developing an R package for dynamic and web-based updating of reviews

ID: 

1140

Session: 

Poster session 1 Wednesday: Evidence production and synthesis

Date: 

Wednesday 13 September 2017 - 12:30 to 14:00

Location: 

All authors in correct order:

O'Connor A1, da Silva N1, Hofmann H1
1 Iowa State University, USA
Presenting author and contact person

Presenting author:

Annette O'Connor

Contact person:

Abstract text
Background:Updating reviews can be a laborious process when it includes updating all the data from meta-analyses into the text, tables and figures included in an original report. Further, static presentation of reviews in print publications limits the access of end users to all the available meta-analysis and the evidence.

Objectives: The objective of this project was to develop a web-based tool that could be used to facilitate updating a review, and in particular, updating the output from R meta-analysis packages into the updated written report. A second objective is to enable presentation of the review (text and any meta-analysis) online in an interactive interface.

Methods: To achieve the objectives, we make use of the tools for reproducible and dynamic reports already available in R such as knitR or markdown. We also use Shiny to create a web-based interface for writing the dynamic reports, updating the meta-analysis results and displaying the results. while remaining user friendly. The package makes use of output from current R meta-analysis packages, and is flexible enough to read different output models from several meta-analysis packages in R. The use of Shiny enables us to organise the review information and display any primary data and meta-analysis results in a user-driven interactive dash board.

Results:The package is currently in development and will be presented at the Summit.

Conclusions: The main advantage of this R package will be to enable a ready access to a modern web-based application for conducting and updating a systematic review for any topic or team, rather than independent of group-specific review production tools.