What proportion of top-shelf subscription journal articles end up in PMC?
In searching PMC recently, I was finding significant numbers of articles deposited from very notable journals, the outcome of results from NIH funded research. What proportion of annual research articles from any one journal end up on PMC? Because the NIH Open Access Policy gives authors at least 12 months to comply and deposit, I thought to explore articles from the year 2009, because we are now well beyond the 12 month deposit requirement, even for articles published in late 2009. Here are my findings:
PMC Deposits as a Percentage of 2009 Article Publishing in Select Journals
| JOURNAL | 2009 Articles (SCImago) | 2009 PMC Articles | PMC Percentage (%) |
|---|---|---|---|
| Nature Genetics | 282 | 117 | 41% |
| New England Journal of Medicine | 1817 | 81 | 4% |
| Trends in Molecular Medicine | 60 | 11 | 18% |
| Journal of Clinical Oncology | 1290 | 288 | 22% |
| JAMA | 1297 | 78 | 6% |
| Cancer Research | 1250 | 577 | 46% |
| Circulation | 1073 | 329 | 31% |
| Diabetes | 388 | 371 | 96% |
| Lancet, The | 1729 | 49 | 3% |
| Gastroenterology | 760 | 121 | 16% |
Some publishers help authors by handling the deposit. Some don’t. The percentages are so small for NEJM and JAMA that one has to wonder whether there is a culture of non-compliance by certain journals. Contrast this mixed bag of compliance with the 1180 journals (as of August 12, 2011) that actively collaborate with the NCBI to make the final published version of all NIH-funded articles available in PMC no later than 12 months after publication, not leaving anything to chance. Actually, the Journal of Clinical Oncology and Diabetes are in this group of public access collaborators.
Read some comments collected by SPARC on the 3rd anniversary of the NIH public access policy.
Wouldn’t you need to know what percentage of articles in NEJM or JAMA are the result of NIH funding before suggesting a “culture of non-compliance by certain journals”?
To echo T Scott’s sentiment, NIH policy compliance is required of authors and institutions, not the journals. Both NEJM and JAMA allow authors to submit to PubMed Central if necessary, but it’s up to the authors to do it.
One more bit of follow-up on this. If you search Pubmed for all articles from NEJM in 2009, you get 1486. Only 112 of those are tagged as having been funded by NIH. If your 81 articles figure is correct, that gives NEJM a compliance rate of 72%.