Advice for improving the reproducibility of data extraction in meta-analysis

Extracting data from studies is the norm in meta-analyses, enabling researchers to generate effect sizes when raw data are otherwise not available. While there has been a general push for increased reproducibility in meta-analysis, the transparency and reproducibility of the data extraction phase is still lagging behind. Unfortunately, there is little guidance of how to make this process more transparent and shareable. To address this, we provide several steps to help increase the reproducibility of data extraction in meta-analysis. We also provide suggestions of R software that can further help with reproducible data policies: the shinyDigitise and juicr packages. Adopting the guiding principles listed here and using the appropriate software will provide a more transparent form of data extraction in meta-analyses.