I analyzed a random sample of 1,957 meta-analysis full-text research papers, uploaded to PubMed Central between the years 2016 and 2021, in order to check the popularity packages of meta-analysis software among medical researchers. (I used the BioC API to download the articles — see the References section below).
Out of these 1,957 meta-analysis papers, only 1,321 mentioned the use of at least 1 meta-analysis software.
Here’s a summary of the key findings
1- Stata was the most used meta-analysis software over the past 6 years, mentioned in 40.27% of papers, followed by RevMan (35.5%) and R (21.88%).
2- The 6-year trend showed that, since 2020, RevMan became the most popular software, beating Stata by having the largest increase in popularity over that period (+1.21%) and taking advantage of Stata’s big drop in popularity (-3.34%).
3- The data also suggest that both Stata and RevMan are more popular among beginners. This is in contrast to R and SPSS, which were mentioned more commonly in papers published in high-impact journals.
Top meta-analysis software over the years
Here are the top 6 meta-analysis software rankings over the past 6 years:
The graph shows that RevMan became the most popular meta-analysis software since 2020 by replacing Stata. Another noticeable trend is the increase of popularity of Microsoft Excel and the decline of SPSS since 2019.
Most popular meta-analysis software overall
The table below ranks the popularity of 24 meta-analysis software packages according to the number of mentions in our sample:
|Rank||Software||Number of Mentions|
(Total: 1,321 articles)
|15||Forest Plot Generator||1||0.08%||–|
⚠ How was the trend calculated?
The 6-year trend is the linear regression coefficient (reported in percent) obtained by regressing “the percent of articles that mention a particular software package each year” onto the “years” variable. This trend was calculated only for meta-analysis software packages with more than 90 mentions over the past 6 years, because otherwise, this number will be reflecting the noise more than the trend.
The data show that Stata was the most popular meta-analysis software over the past 6 years, used in 532 out of 1,321 meta-analysis papers (= 40.27%). However, its popularity is declining, losing 3.34% of mentions each year.
Do beginners and professional scientists use the same meta-analysis software packages?
In order to answer this question, I compared software packages used in articles published in low versus high impact journals.
I collected the journal impact factor for 1,108 of articles in the sample, and divided the dataset into 2 parts:
- Meta-analyses published in low-impact journals (impact factor ≤ 3): This subset consisted of 684 articles.
- Meta-analyses published in high-impact journals (impact factor > 3): This subset consisted of 424 articles.
I chose the threshold of 3 for no particular reason other than it seamed a reasonable limit, and also separates the dataset into 2 large subsets of articles.
The results were as follows:
Looking at this bar chart, we can conclude that Stata, RevMan and Excel tend to be more popular among beginners and the inverse is true for R and SPSS.
Meta-analysis software packages frequently used together
Out of a total of 1,321 articles in our dataset, only 433 (32.8%) reported the use of more than 1 meta-analysis software.
Here’s a table that shows the top 10 pairs of meta-analysis software packages frequently used together:
|Pairs||Number of Mentions|
(Total: 1,321 articles)
|Stata + RevMan||124||9.39%|
|Stata + R||44||3.33%|
|Stata + Excel||44||3.33%|
|Stata + SPSS||26||1.97%|
|R + RevMan||21||1.59%|
|Stata + SAS||18||1.36%|
|RevMan + Excel||18||1.36%|
|R + SAS||17||1.29%|
|R + SPSS||15||1.14%|
|R + Excel||14||1.06%|
I don’t think there is much to interpret here, as the top 3 most used meta-analysis software packages are also the ones frequently used together.
An important thing to note is that Microsoft Excel was the least used software on its own; it tends to be used with other packages in 66% of cases, compared to only 32% for R — which was the most software package used on its own.
- Comeau DC, Wei CH, Islamaj Doğan R, and Lu Z. PMC text mining subset in BioC: about 3 million full text articles and growing, Bioinformatics, btz070, 2019.