Bibliometric Data Analysis Scripts

Instructions for Running Data Analysis

1. Clone this repository in the Rstudio.

2. Install the requeried R packages (R version >4.2):

   • rmarkdown
   • tidyverse
   • rlang
   • lubridate
   • Hmisc
   • compareGroups
   • flextable
   • table1
   • wordcloud2
   • stringi
   • webshot2
   • webshot
   • htmlwidgets
   • maps
   • countrycode
   • bibtex
   • officer

3. Execute render.R script from RStudio

File descriptions

• /scripts : Includes R file scripts render.R and analisi_StateArt.Rmd (main script), and R functions.
• /data: Datasets downloaded from PubMed (papers.txt and objects.txt) and Countries.txt.
• /outputs: All the outputs generated by scripts are stored in this directory.
• conductor_state_of_art.xls: Description of datasets and metaparameters required to execute scripts.

Analysis summary

For bibliometric and visual analysis, we included all retrieved articles that met the previous criteria. The following bibliometric indicators were extracted from each paper and imported for in-depth analyses: publication year, journal name, journal country, abstract, keywords, impact factor, quartiles, and categories. A descriptive analysis was conducted to examine the frequencies of these variables. Additionally, we performed graphical analyses, generated a Word cloud illustrating the distribution of keywords and mesh terms, and assessed frequency patterns based on the region and country of the first author. We identified variables related to the study design and evaluation domains through word searches in the title, abstract, and keywords. A list of synonymous words used for design and domain categorisation.