PubMed sucks, or the user is broken

Anna Kushnir runs a blog on a high profile platform over at Nature Publishing. Last Saturday she complained about the user fiendliness of PubMed.

I have spent an absurd amount of time on PubMed recently and can say in no uncertain terms that it is making my dissertation writing way more painful than it needs to be. I can hold a paper in my hands, search for two authors’ last names and have PubMed come up with nothing.

PubMed, however is probably the most widely used bibliographic database in the world. Certainly in the world of Medicine. Many libraries run special classes to teach the intricacies of PubMed. We -librarians- have to admit, searching PubMed is not easy.  It is certainly not intuitive. After you’ve found what you searched for, then it is complicated to get the information over to another programme such as Reference Manager or EdnNote. If you succeed in that, you get abbreviated journal title’s, authors with maximally two initials etc….

How surprising was the reaction of Dean Giustini. Well his reaction is perhaps typical for a librarian in general, we go out and teach the user a few tricks. We teach and teach. The database is not broken! It’s the user we need to mend.

I thought Dean would know better than this. Of course he is right in the fact that this complaint on PubMed is an excellent teaching moment. But I would rather stress the message from Anna Kushnir, and that is that searching PubMed is not intuitive. Far from it. Even if you would have had classes some years ago in searching PubMed, that knowledge is now obsolete. That is good for PubMed, they innovate and improve, but when we think that refresher courses in searching PubMed should be high of the lists of Doctors, surgeon and medical researchers, we are speculating on the wrong track. They simply don’t have time for these courses. It is a rat race to keep informed on the progress of their own specialities. Why would they need courses for full time MLIS professionals to search a bibliographic database?

We have to go out there and listen to our users. Anna Kushnir is one of them. Her message is plain and simple, searching PubMed -however good we think it already might be- should become more intuitive. I think we should do a lot better and can do a lot better to build these more intuitive search engines.

I see the post from Anna more as a challenge for our profession, than as a teaching moment.

11 thoughts on “PubMed sucks, or the user is broken”

  1. Hi Wow!Wouter,

    Great post. I agree it’s a challenge for us professionally to deal with PubMed both in the short and long term. Would you agree with the idea that some of the questions Anna has could be answered quickly and that this would at least alleviate some of her apprehension as she completes her research?

    With best wishes,


  2. Hi Dean, Yes it is a great teaching oppurtunity. That is true. But we only address so few. For the long term we need to mend PubMed.

  3. Hello, WoW!ter. I think you have hit the nail on the head, in a much more eloquent manner than I was able to in my somewhat angry rant against PubMed. The site is just not intuitive, nor is it user friendly. Today I took the time to look at the help page. There is so much material there! There is no way that I can make the time and summon the attention span to make my way through all of it, nor do I think I should have to. There are plenty of search engines and sites that are easier to use than PubMed. I am loyal to the PubMed brand, however. I would like to continue using PubMed and I want to see it improve and grow. I very much appreciate you taking note of my post

  4. The way I try to pitch it when I’m working with students who are starting to think like researchers, PubMed and its ilk are tools in the research toolbox, and there are many tools used in biomedical research that are not completely user-friendly. Dissection, cell culture, PCR, electron microscopy etc – nobody just sits down at the bench and expects those to be intuitive. But because PubMed is on the web we have these expectations that it will be as easy as shopping.

    I’m not suggesting that it’s a perfect interface or that NLM shouldn’t try to improve it. It was pushed onto the web in response to political demands and the PubMed interface wasn’t really designed for researchers (ask an old-timer about the old superior Internet Grateful Med interface – sniff, sniff). But it seems to me that in its current state, it’s probably worth the half-hour it will take to sit down with a librarian and get you up and running with alerts and whatever other feature you need. Getting local help (as opposed to from NLM’s help site) is also useful when it comes to your own institutional route to the e-journals and how to request stuff. Anyway, just tossing in my two cents. I do get the frustration – I see it all the time. It’s frustrating! :)

  5. Isn’t Anna Kushnir just turning things around and looking at things backward? Medical terminology and publishing methods weren’t thought up by information professionals, though we do our best to keep track of them! You can’t keep all of the people happy all of the time, but maybe one way to enrich PubMed would be to allow authors to tag their own articles with ‘natural language’ and current jargon tags. This could be a help for the Google generation.

  6. PubMed is broken. I have had students complain to me that “it only has one search box.” The traditional librarian response is “teach the user — they should learn it OUR way! We are right! The user is wrong!” (or “the user is broken” — hee)

    But in this Google / information overload age, that no longer holds any weight, at least in my opinion.

    PubMed NEEDS to be more user-friendly. Researchers, students, medical professionals — their goal with PubMed is to FIND research and then study it. Their focus should be on the finding, assessing results for relevance, retrieving the full-text of the citation they’ve identified, and then be gone from PubMed.

    Researchers, students, etc. SHOULD NOT have to be expert searchers in order to find what they need. That’s unrealistic and a misunderstanding of their role. Searching be seamless for them, simply a tool, like scissors.

    Our role as librarians is to advocate for users with vendors. Unfortunately, I’ve had several go-rounds with PubMed on user interface issues (mainly to get our OpenURL resolver to work properly in PubMed so that users can get to the full-text) and have received only perfunctory answers from them.

    It’s a serious problem. Great content that’s almost impossible to get to unless you’re a medical librarian.

  7. Guys I certainly agree with this, infact Iam a hardcore molecular biologist who was finding it tough on PUBMED, Now I have shifted to xtractor- a free portal which makes searching simple and gives you the most accurate data since its manually curated by humans..
    has rich features on community building as well

Comments are closed.