PubMed Commons to be Discontinued

This post was updated on February 27, 2018.

Update: NLM appreciates all of the input we have received in response to our February 1, 2018, announcement that PubMed Commons is being discontinued. Thank you for your thoughtful comments. We are heartened to hear that many of you found it to be a useful service.

As we mentioned in the announcement, comments on articles indexed in PubMed will continue to be visible on PubMed and PubMed Commons through March 2, 2018, after which time they will be available for download from NCBI’s website. NLM again thanks all of you who participated in PubMed Commons for your interest and effort.

PubMed Commons has been a valuable experiment in supporting discussion of published scientific literature. The service was first introduced as a pilot project in the fall of 2013 and was reviewed in 2015. Despite low levels of use at that time, NIH decided to extend the effort for another year or two in hopes that participation would increase. Unfortunately, usage has remained minimal, with comments submitted on only 6,000 of the 28 million articles indexed in PubMed.

While many worthwhile comments were made through the service during its 4 years of operation, NIH has decided that the low level of participation does not warrant continued investment in the project, particularly given the availability of other commenting venues.

The discontinuation plan is as follows:

  • New comments will be accepted through February 15, 2018.
  • Comments will continue to be visible on the PubMed and PubMed Commons websites through March 2, 2018.
  • Users wishing to access the comments after March 2, 2018, will be able to download them from NCBI’s website.

Many thanks to all of you who participated in this experimental effort to enhance the opportunities for interaction about published biomedical literature.

76 thoughts on “PubMed Commons to be Discontinued

    1. This statement is not quite accurate. For example, PubPeer comments are mirrored on web sites of some journals (AND PubMed Commons) but not on the others. The problem is rather that some journals, as well as authors, do not accept any post-publication peer review.

      1. Well, I asked that very question at the OASPA meeting last year (speaker was Hilda Bastian, Editor of PubMed Commons), on whether an API exists or is planned, and the answer was “no”. Many journals do have some sort of commenting functionality, so it is not that journals reject post-publication peer-review. Our journals (JMIR) even display tweetations (citations in tweets, when user comment on an article via Twitter). We would have gladly shown PubMed Commons comments if an API would have been available.

  1. What are the other venues for commenting that NCBI has decided to endorse in lieu of the verified, neutral, publicly funded PubMed Commons? And do these venues offer greater value?

  2. This is a pity. It was an opportunity to add perspective and constructive critique. Engagement was certainly low, but this is likely because it lacked promotion. It also wasn’t really PPPR, but, as noted in the notice, a place for discussion, easy addition of updates or cross-referencing. Publishing is going to significantly change for the better in the next decade and such experiments will find traction, I’m sure.

    Here’s hoping PubMed will not list clearly predatory journals.

  3. This is extremely disappointing – I would prefer to see this improved rather than abandoned. I added a substantial number of comments and I think/hope they are valuable – I often used them as leverage to promote this activity with young scholars as a public and traceable way to participate in the dialog, address sloppy practices by scientists and publishers… so many things need more and not less crowd-sourced content.

  4. This is a shame, as PubMed Commons provided a stable and easily accessible venue for comments and discussion. As many journals do not publish Letters to the Editor commenting on their publications, the PubMed Commons venue plays an important role. For example, we wrote a Letter to the Editor about some mistakes in a manuscript, and the journal told us to instead submit it as a Comment on their web page for that manuscript. We did so, but their web site was so poorly designed that the only way readers visiting that web page would see the comment would be if they actively clicked on the ‘Comment’ tab (which provided no indicator at all that it contained a comment). In contrast, our comment on PubMed Commons was very visible and accessible to all who accessed the PubMed web page for that manuscript.

    This is a loss to the scientific community.

    1. Agree, it would be a great loss. Hopefully NCBI will rethink. The unusual formatting requirements did not help, and the quotation marks in pastes from my MS word files had to be adjusted. But the PMIDs provided easy access to the literatures cited,

  5. Starting with my first comment in February 2014, I posted 234 carefully-written comments on PubMed Commons. I believed that PubMed Commons would be the future of medical publishing, a channel of discussion that bypasses overworked journal editors. PubMed Commons was a great idea that was too early for the future and too futuristic for the present. Enjoy your letters to the editor with 1 year turn-around!

  6. Dear PubMed Commons,

    We just read the distressing news that PubMed Commons will be discontinued. Discontinuing this valuable “forum for scientific discourse” is allegedly justified by the availability of “other commenting venues”. This justification is inaccurate because there are no other commenting venues that are readily available to published authors and free of editorial bias and suppression of dissenting views.

    PubMed Commons provides a unique resource for commenting on publications that have passed editorial review despite significant flaws. Under these circumstances, journal editors are reluctant to expose those flaws and generally suppress critical comments such as Letters to the Editor about the published work. PubMed Commons provides a means to set the scientific record straight and bypass politically-motivated editorial suppression of these important comments.

    We urge you to continue PubMed Commons. This valuable scientific resource should not be eliminated.


    Raphael B. Stricker, MD
    Lorraine Johnson, JD, MBA

  7. This is a valuable resource and should not be discontinued. The reason it is underutilized is that it was not adequately publicized and promoted. Please do not discontinue Pubmed Comments 🙁

  8. The medical profession, including medical schools and hospitals, is now a part of the health care industry, and implementation of editorial policies of medical journals is commonly biased in favor of business interests. PubMed Commons has offered the only, albeit constrained, open forum to air dissenting research and opinions in science-based language. Discontinuation of PubMed Commons will silence any questioning of the industry-sponsored promotional publications indexed in PubMed.

    1. Just for the record. After I posted this comment to several Abstracts under which I posted comments previously, my PubMed Commons account is suspended.

  9. Being a faithful participant of PubMed Commons, I am among many others disappointed by the news. I did not truly consider PMC as an experimental effort to enhance the opportunities for interaction about published biomedical literature, as stated by NCBI. I considered PMC as a great opportunity to deal with the breaking down of the peer review system, and/or give what I expected to be helpful suggestions. I was expecting some feedback, it barely happened. Browsing other posts however made me think it was moving extremely slowly but in the right direction. Discontinuing it goes against the advancement of science, it is going backwards instead of forwards. Actually it could have evolved into a system designed to hold both authors/editors accountable for not addressing/correcting serious flaws in their published works. Should not we all be held accountable to the next generation of scientists? NCBI admits many worthwhile comments were made but they decided to remove the comments from PubMed. Why? Is there any respect left for the dedicated caring scientists?

  10. PubMed Commons is invaluable in my field, myalgic encephalomyelitis/ chronic fatigue syndrome. ME/CFS is a chronic, disabling medical condition that affects at least a million Americans and for which there is no effective treatment currently. Despite efforts over the last decade by the US CDC, NIH, FDA, Institute of Medicine, and some universities (Stanford, Columbia, Cornell, etc.) information takes a long time to spread and many scientist and physicians continue to inaccurately perceive it to be a personality disorders and/or psychological/ psychiatric condition. This includes journal editors who often do not publish dissenting letters or letters that provide an alternative perspective, despite those letters coming from prominent investigators at well-known institutions.

    (Not specific to ME/CFS, it is well known that journals often shut out dissenting or revolutionary ideas out of prejudice rather than merely poor adherence to experimental priniciples. Nobel-prize-award winning manuscripts have been shown to be rejected by multiple journals initially. Hence, Planck’s statement that science advances one funeral at a time.)

    PubMed Commons allowed me and others to comment on those studies without the prejudiced refereeing by some editors. Additionally, people who don’t personally subscribe to specific journals or are affiliated with medical libraries often do have access to letters to the editor but they do have access to comments on PubMed Commons. Finally, the audience Commons reaches is likely wider than many journals as readers stumble across comments in their searches on a specific project. Like the other commenters here, I hope PubMed Commons returns or is replaced with something similar. Perhaps PubMed needs to better publicize PubMed Common’s advantages over the usual Letter to the Editor submission process.

  11. I can see why this would be a tough call for PubMed and I can certainly see why some are urging the NLM to reconsider. PM and PMC perform a unique, invaluable service to humankind. Their accomplishments are immense and their integrity is appreciated. A shining example of what government does and can do.

    Still, citations (versus full text) are not particularly well suited for robust discourse. Here I’m thinking of newer tools such as Hypothesis that enable inline annotations and discussions. And, as the announcement eludes to, discussions are occurring elsewhere on platforms optimized for such activity. The uptake is feeble given the prominent real estate on PubMed’s high traffic home page. If PubMed went “all-in” to enhance the service, it may gain traction over time, but this might require a potential realigning of mission or repurposing resources from other initiatives (conjecture, I realize).

    Those who think that poor usage could have been bolstered by APIs or publisher partnerships don’t appreciate the complexities of the commercial publishing landscape and what drives these publishers. There are exceptions, of course, but the content earns billions for the publishers and there’s no magic wand to win buy-in and adoption.

  12. I echo the comments of previous contributors. I found PubMed Commons to be a useful source of information and debate, with a particular strength being that it linked directly to an indexing service used by the vast majority of scientists almost daily. I agree that the lack of usage was more to do with a lack of publicity and awareness within the scientific community than to do with nonchalance amongst researchers. With a determined drive to publicise the service, I believe it could rival, and potentially out-compete, the alternative commenting venues referred to above.


    Very disappointing. The rationale is strange: “comments submitted on only 6,000 of the 28 million articles indexed in PubMed.” Given the effort required from the busy people who make important comments, and the massive number of articles (many not worthy of comment), 6,000 is a remarkably good result. I have been pointing out the value of PubMed Commons on the Scholarly Kitchen blog. These are early days in a very important innovation. It should not be dismissed as a failed experiment.

  14. Suppressing commenting opportunity would definitely be a loss to the construction of scientific knowledge. There are not that many commenting alternative, as commenting is not always allowed on journals’ websites and many other commenting tools are only included in closed networks and thus not publicly visible without an account.

  15. What will happen to the permalinks when PMC shuts down? Will they still bring up comments in PubMed so people can use them as references , and if not, could they be set to forward to PubPeer’s copy?

  16. I also will be greatly disappointed to see pubmed commons go. It is a great venue for discussing points about others’ papers as well as presenting corrections/updates to my own. Since the system is already in place, it can’t be be too expensive to maintain. One should not feel embarrassed that the vast majority of papers have no comments, the value is in the substantial number that do. Some papers are very straightforward and specialized, and may not suggest any comments. Future results an publications may inspire updates and corrections and pointers to follow-up publications, long after the original article appeared.

  17. This is very disappointing. It was a wonderful idea, but was not well known in the research community. It needed a longer trial and greater promotion to get the usage up.

  18. Please reconsider this decision. This is a valuable beacon pointing at where scientific publishing needs to go, if for no other reason than to provide a place for questions to be raised about techniques or results in even a small fraction of papers. Small usage is not a good reason to kill it now.

  19. Last week’s decision to close PubMed Commons is a sad turn of events. It takes a great deal of effort in our hectic research lives to analyze and compose comments on articles. I certainly do not understand why the NCBI has decided to remove all of the comments that have been posted hitherto. Why not leave them in place? The people who have posted these comments have spent time analyzing and evaluating individual articles and in the cases when there are serious issues, they have taken the time to compose warnings about possible problems. These comments can be useful for researchers who wish to build on a published study. Why leave the original articles in PubMed, but remove the potential warnings about these articles?

    Publishers of articles, which turn out to have issues and for which comments have been posted on PubMed Commons or even for which corrigenda or errata may have been issued, often do not make these corrigenda and errata readily visible or not visible at all on their websites. In some cases, the corrigenda may be visible on the journal website, but for some reason are not seen on the PubMed entry. I have heard of some journals making it difficult for authors to issue corrigenda for their articles. We clearly need a fix of this situation, but in the meantime, PubMed Commons is one way of warning the reader that there is an issue.

    As an illustration from my field of cell line authenticity, there are many articles which should have been retracted or, at the very least, corrigenda issued. However, very few corrigenda are actually issued or articles retracted. To date, I have been able to identify about 20 articles retracted because of cell line identity issues. Instead, authors publish the first article about the cell line they have established. Then use the cell line in several articles and share it with other researchers who use it to publish more articles. Years later, the first group (or some one else) discovers the cell line is not what they claimed and publishes an article that they have found that this “new cell line” was contaminated and was, for example, HeLa or some other culprit. Unfortunately, this correcting article is neither linked to the various articles based on this imposter cell line nor are corrigenda linked to the various articles that were published using this imposter cell line on the websites of the journals. Thus a problematic cell line leaves an “unblemished track record” of publications in the literature.

    I have several questions for the NCBI leadership.
     How is the uninformed researcher going distinguish the useful solid literature on which they can build their research from the shaky research on which they will waste their efforts and research funds? PubMed Commons may not be perfect, but it is one of the few methods of ameliorating this problem.
     PubMed Commons may not have been used as much as the NCBI leadership had hoped, but why tear down what it has built by removing the already contributed comments?
     Is the NCBI going to offer some alternative to these mechanisms for addressing problems in the scientific literature when today the reproducibility of much scientific literature is in question?

    Leaving thoughtful comments about the literature is a thankless endeavor, but those who take the time to do this work are providing necessary feedback to both individual research groups and the scientific community. I urge the NCBI leadership to reconsider this decision to close PubMed Commons and remove the comments already posted. At least, PLEASE leave the comments in place!

    1. Could not agree more. And like another contributor above, I have been advised by an editor of a journal that does not have a commenting facility, to place my concerns in PubMed Commons.

  20. PubMed Commons should be continued, with full financial support provided by the publishers of PubMed-indexed journals. PubMed indexing is a valuable tool provided to publishers by NCBI, so let the publishers support continued commenting. The publishers can select from the posted comments to choose the best for publication in the journal itself, along with any author replies posted. How can publishers be compelled to pay for PubMed Commons operating expenses? Fold these expenses into the cost of running PubMed, and charge each publisher for each PubMed indexed abstract posted, along with its PubMed Commons comments.

  21. 1) Please provide instructions for downloading PubMed Commons without the yellow termination notice at the top of each page.

    2) My PubMed Commons comments fill up 12 web pages, with about 20 comments per web page. Is it possible to download all 12 of my web pages into one file, rather than 12 separate files?

  22. It is sad to hear that this post-publication peer review project will be discontinued. This is a real loss to the community of researches who are worried about the broken scientific publication system. When I first heard about PubMed Commons, I found it to be one of the few true innovations in scholarly communication. I can only echo what other people here have commented: Many journals do not offer the possibility to submit a short commentary. PubMed Commons was perfect for that and easy to use.

    6000 comments on 28 million articles. I think that is brilliant and should have been put in a different way: 6000 on how many records since the service started? Now that the service is implemented: How much resources does it need to keep the service running? Can´t be that much!

    I hope NLM considers recalling this decision.

  23. Please continue, as this is a very valuable platform for scientific discussion and an efficient way for pointing out some errors in the published papers

  24. PubMed Commons was started as an experiment and the way it turned out was that the kind of comments made was mainly post-publication peer review usually hinting at (potentially) severe problems with the indexed articles. This is incredibly helpful! General agreement seems to be that the quality of comments is generally very high. In my teaching of medical students PubMed Commons was one of the sources to refer to for critical appraisal and for a final check of cited articles.

    The posts above and many of the comments in PubMed Commons itself show that a lot of PubMed authors consider PubMed Commons a very valuable instrument to remedy problems arsing from a broken journal publication system.

    To add another example: In one of my articles the journal’s editor refused to publish an electronic supplement containing methodological details necessary according to reporting guidelines cited in the paper. The supplement would be “over the head of the readers” of that journal. As this happened during the last step of the review process after the paper was accepted it was too late to submit to some other journal. Via PubMed Commons I was least able to provide a link to the supplementary data in our repository.

    Comments can only be made by authors registered in PubMed and very often after an additional invitation by some other author already admitted to PubMed Commons. Not always so easy to find somebody to let you in. So there is quite some entry barrier. And then comments will be publicly linked to an author’s personal profile and to all their papers in PubMed. In my expectation it does take some guts and considerations to publicly criticise an article under these conditions. This might not necessarily be one of the things to do to boost your career.

    Authors did comment under the assumption that their comments will be visible with an article (This is what PubMed is good for) and for a long time (This is what you expect from NLM). At least I did. Sorry to be disappointed.

    I am very sad about the decision to discontinue PubMed Commons and to even remove the comments from PubMed. The argument for this decision was that the number of comments, “only 6,000 of the 28 million articles indexed in PubMed” would be too low to further justify the costs incurred. I do not think the argument with numbers is convincing.

    There are other means in PubMed that hint at (possibly) problematic articles. Let’s take a look at these by searching PubMed:

    Retracted Publication [PT] –> 5700 records
    Retraction of Publication [PT] –> 5875 records
    hasexpressionofconcernfor –> 195 records

    These article counts (out of a total of 28 millions) are about “as low” as those with PubMed Commons comments. Should we be happy or sad? Are these numbers low or are they high? When compared to what? There are probably costs involved in curating these data. Should the system to mark retracted and other problematic articles be abolished? Or is this important as a means of quality control?

    May I ask NLM to reconsider the decision to discontinue PubMed Commons?

  25. Regarding original material I posted to PubMed Commons and nowhere else, I have been citing the URL of my comment on PubMed Commons on subsequent PubMed-indexed publications I have written. Will readers be able to click on those URLs see my original comment on PubMed Commons?

  26. Another useful aspect of PubMed Common’s linkage to the NCBI database was that it greatly facilitated the peer-reviewing of papers, grant applications and recommendations for high awards. Each of these has a list of author publications. One could paste titles at the top of the PubMed Commons page and with a few clicks see if there were any comments on each paper.

  27. I would vote to continue PubMed Commons if I could. There are a lot of useful comments which include good critics, added some great links to other related publications or just added more information to some publications.

Leave a Reply