PubMed’s Backdoor Makes Me Question Quality

Ok, I am dancing in my seat saying “I told you there is a problem with PMC as the backdoor to PubMed.” I know it is gloating but librarians get so little to gloat about so forgive me.

Back in 2011 I wrote the post, Backdoor Method to Getting Articles into PubMed: Is indexing so important?  At the time I was more concerned about the findability and lack of indexing of PMC articles that found their way into PubMed.  It wasn’t until a few years later with the popularity of Beall’s List did I begin to think about the quality of PMC articles now in PubMed.  In a discussion on the Medlib-l listserv regarding Beall’s list I mentioned the PMC backdoor again.  As I said, ther are “some researchers who see no distinction between PMC submitted journal articles from non-indexed journals and ones that are indexed in MEDLINE. To them it is in PubMed and that is good enough.”

I am so happy that Kent Anderson in the Scholarly Kitchen wrote the post, “A Confusion of Journals – What is PubMed Now?”  His post along with articles and communications published in peer reviewed journals Neuroscience, Lancet, and Archives of Physical Medicine and Rehabilitation mention the ever growing problems of predatory journals within the profession.  Neuroscience and Lancet specifically mention the problem within PubMed.

Unfortunately, many doctors and researchers believe what Jocalyn Clark and Richard Smith write in the editorial piece in The BMJ, “Articles in predatory journals, although publicly available through internet searches, are not indexed in reputable library systems. The articles are not discoverable through standard searches, and experienced readers and systematic reviewers will be wary of citing anything from these journals. The research is thus lost. ”

That is a partially true statement. The articles in PubMed Central are NOT indexed in MEDLINE, HOWEVER they are totally discoverable through standard searches in PubMed.  The “research” from that predatory journal is not lost.  It is is very discoverable in PubMed.

Personally, I think this is a mess.  NLM should not have given its repository the name PubMed Central, it is confusing and just the name muddies the waters.  But that ship has sailed.  NLM needs to put some serious authority control on PMC quickly.  Not only do articles in predatory journals water down what is already indexed in MEDLINE and within PubMed but it could be have significant problems for the treatment of patients and researchers.  Fake news seems to be the phrase of the day recently.  More than ever people need reliable resources where they can find credible information.  Allowing articles from predatory journals into PubMed through PMC creates a credibility problem with PubMed.  It is basically like seeing Gweneth Paltrow’s Goop medical advice published on WebMD.

I really hope Kent’s post and the other researcher’s articles get the attention they deserve.  I hope it leads others to look at and question the quality of articles allowed into PubMed by virtue of PMC.  Then perhaps NLM will look at quality control methods for PMC.  If they don’t they run the risk of ruining PubMed.  We already use another (pay) database when we must assess the quality of article.

6 thoughts on “PubMed’s Backdoor Makes Me Question Quality”

  1. Thank you! There have been many bad decisions regarding PubMed and branding over the past years. Besides PubMed/PubMed Central, there is the additionally confusing “MedlinePlus” which has absolutely no relation to Medline.

    NLM, in their apparent efforts to make PubMed more accessible to non-librarians, have steadily removed or hidden important information that we — the constant users of PubMed — use to do our daily work, train others, and support medical research. Examples of this are legion.

    For instance, the removal of the “In Process” and “As Supplied by Publisher” tags from public display, which are now only viewable in the MEDLINE or XML formats (https://www.nlm.nih.gov/pubs/techbull/ma17/ma17_pm_update.html). These tags were extremely useful when training users to distinguish between indexed and non-indexed articles quickly.

    Replacing the “Limits page” by a results filter sidebar and subsequently hiding some of the most important filters (such as Language and Ages) behind “Choose additional filters” or “Customize” links further increased not only the effort needed to modify search results, but made them essentially invisible to casual users, not to mention the increased difficulty in teaching the resource (https://www.nlm.nih.gov/pubs/techbull/mj12/mj12_pm_sidebar.html).

    The new format required at minimum a 4-step process to apply a now hidden filter: click “Choose additional filters,” click the category (e.g., “Ages”), click “Show” (Previously “Apply”), and finally click the desired filter to modify the existing results. Filters further hidden behind a “Customize” link require an additional step!

    In July of 2106, the incomprehensible decision was made to move “The ‘Items per page’ selection … from the top of the results page because it is infrequently used by searchers” (https://www.nlm.nih.gov/pubs/techbull/mj16/mj16_pubmed_display_changes.html). I’m not sure which “searchers” they are talking to but I have used it often.

    Why leave other similar functions (Format, Sort by) and remove something at least as useful? Fortunately this bad decision was reversed less than a year later (https://www.nlm.nih.gov/pubs/techbull/ma17/ma17_pm_update.html).

    Removal of the “History” link from the PubMed search page served to hide this feature from uninitiated PubMed users (https://www.nlm.nih.gov/pubs/techbull/nd08/nd08_pm_news.html).

    I could go on and on, but I think I’ve made my point. NLM/NCBI’s constant monkeying with PubMed and its interface makes life for users and trainers more difficult.

    Back in June 30, 2000, a Technical Bulletin titled “The Transition to New PubMed” was published and contained the following statement “… the changes we make to PubMed are the result of feedback and requests we receive from PubMed users.” I have to say, I find this difficult to believe.

  2. I agree with Krafty Librarian! Yet, this issue does provide an opportunity for us Informationistas to continuously reinforce and teach our readers/scientists/researchers how to identify bogus/predatory content. We can speak out about it! We can incorporate a sort of “predatory indicators” in our work. A list of predacious signs to watch for as one reads. We can incorporate these “predatory indicators” into our daily work, websites, classes, webinars, blog posts. After 18 years of systematic review work in the VA Technology Assessment Program, I can attest to the fact that there are plenty of bad studies reported in good journals; good research published in not so great journals, and some outright “crap” published in the un-indexed journals along with the good research published there. It’s up to us–the knowledge professionals to sound the clarion call for “reader beware” acknowledging that we likely can’t stop the predatory publishers – but we can take a YELP approach and “out” the crap when and where we see it. We can comment on PubMed now, we can raise awareness in the arenas we work in, we can blog, rant, podcast, write about what we see, and encourage others to do the same. We can take this opportunity to channel our outrage into raising awareness and enlightening our communities. MLA CE 2-3 hour class? MLA webinar? NN/LM Webinar. . . .

  3. The NLM is holding a free webinar on October 6th entitled, “PubMed Journal Selection and the Changing Landscape of Scholarly Communication.” We will discuss the PMC journal review program, but also journal selection for MEDLINE and the current publishing environment. The webinar is open for anyone to attend.

    You can read the full description of the webinar at the following link: https://nnlm.gov/class/pubmed-journal-selection-and-changing-landscape-scholarly-communication/7780. To sign up, you will need to create an NNLM login by clicking “create new account”.

  4. Remember when the mention [Indexed for MEDLINE] appeared in the Summary display (whereas it only shows up in the Abstract display now)? Well, it was an easy way to find out about the quality of a journal.

Comments are closed.