Webinar Materials

On January 28, 2022, Catherine Staley and a panel of experts from the National Library of Medicine and the Network of the National Library of Medicine hosted NLM Office Hours: PubMed. It includes a conversation with PubMed product staff from the National Center for Biotechnology Information (NCBI).

Links

• PubMed Search Fields Descriptions and Tag from the PubMed User Guide

• User Suggestions for Medical Subject Headings

• PubMed Online Training

• NLM Technical Bulletin

• NLM Support Center

Transcript

>>Okay. It is 12:01:00 p.m. here on the East Coast so we'll get started. We have a little bit of housekeeping to start. Welcome all to our first Inaugural NLM office hours. My name is Mike Davidson, from the Office of Engagement and Training from the National Library of Medicine.

The goal behind these Office Hours sessions is to give you a chance to learn more about NLM products and get your questions answered. We will be recording and posting these sessions so those that are not available at this time will be able to watch them later.

We plan for these office hours to occur regularly, featuring a variety of different presenters, panelists and products. Today we are focusing on PubMed, and we have a great roster of folks to help answer your questions. We have Kathi Canese, program manager for PubMed at the National Center for Biotechnology Information, or NCBI; also joining us from Kathi’s team are Jessica Chan and Amanda Sawyer. We also have Kate Majewski and Catherine Staley from the Office of Engagement and Training. Between all of these incredibly knowledgeable folks, we should be able to answer all of your PubMed questions today.

We've muted the attendees in order to avoid cross talk, but you can put any questions that you have in the chat at any time. I encourage you to send the questions to all of us so we can make sure we can get an answer to them. I'll take a note of those questions that come in and try to direct them to the right expert during our Q&A.

We're going to start out with a brief presentation from Catherine Staley. Catherine, you want to take it away.

>> thanks, Mike. Today I’m going to demonstrate four PubMed features for you. These will be particularly helpful if you're working on a team or systematic review. But first, I would like to take a quick poll to find out what your preferred method is for saving and sharing is. Your options are a citation manager, E-mail, a My NCBI collection, A CSV or Text File, or different method, and if it's different feel free to tell me what that is in the chat.

I'm give everyone a minute to respond. It looks like e-mail is pulling ahead a bit. All right. I think we've got most responses so thank you for your responses and we'll show the results of the poll now. So that you can see where we landed. Yes. So, I’m going to be showing you quite a few of these options today. And I hope that you'll learn a new way to share and save your results that you might be able to use in the future. Thank you for sharing. All right. We can close the poll now. Thanks, Brittney.

For the next few minutes these are the different features that I’m going to demonstrate. Sending results to a citation manager. Saving results as a csv or text file. Saving results to myNCBI collection. And downloading your search histories and details. I'm going to share my screen with you so one moment while I do that. There is PubMed.

So, in the interest of time and because we're focused on saving and sharing today, I’ve already prepared and ran a search that includes multiple terms and filters so that we've got some good results to work with.

And the first thing that you should know about saving your results is that you have some options about what you want to save. And with all of these different saving options that I show you today you can elect to save all of your results, the results on a specific page or a selection of results that had you choose.

* Also, I’ll select a few for this demonstration and notice after I select one and scroll down the page the options appear at the top of the page and they stay there while I scroll. So, that is a recent change that was made to PubMed based on user feedback.

So now you don't have to scroll to the top of the page to use those features, and that brings us to the first option which is to send your results to a citation manager. So, to do that click the Send To button and then select Citation Manager. And then this menu will appear, and you can select the results that you want to include.

I'm going to stick with Selection for today and when you select Create File a .nbib file will automatically download, like you see here at the bottom of my screen. And you can take that file and import it to the citation manager of your choice.

The second option is to save results in a variety of formats depending on the information that you need. So, I will click on Save from that same menu. And walk you through these different options. So, the first format option is Summary Text. And this will download a text file with a list of citations for your results and then you can copy and paste those citations into a Word file or wherever you need them.

Another option -- Abstract Text -- also downloads as a text file but it includes abstracts when they are available with your citations. The CSV File option formats your result citations in a csv file and each part of the citation is in its own column so you can sort by title, author, journal, etc.

Now there are two more saving options here that might not be as familiar to you which is PubMed and PMID. If you select to save as PubMed, a text file of your results in the PubMed format will download and this means that you will see all of the citation elements included in any field, not just the one that appears in the abstract view from the results page.

So when would you use this PubMed format? One scenario is if you’re performing a comprehensive search with MeSH terms that a topic was previously indexed under, and you use MHDA or the MeSH date tag to only include citations that were indexed during a specific range of years. The abstract view in PubMed does not include the year indexed in MEDLINE, but the PubMed format does -- so you can use this saving option to see why a citation is included in your results.

And the last option under the save button is PMID. This option will download a text file with the PMIDs or PubMed identifiers of your results. And you can use those PMIDs to view your results by copying and pasting them back into the PubMed search box and PubMed will automatically OR the PMIDs together. Those are the options that you can find under the save button.

Next, I’m going to demonstrate how to save your results as a myNCBI collection. A myNCBI collection is a great way to access your results from any device or to share them with a team. So, to start, I’m going to go to the top of the PubMed page and log in to myNCBI account.

I'm already logged in so I will skip that step. Once you're signed in, click on the Send To button and then select Collections. Now from here you can choose if you want to create a new collection or add to an existing one. I'm going to click Add to an Existing One. To view a collection or to share it I’ll go back to the myNCBI account and specifically to my Dashboard. And in the Collections box are all of my collections.

If I wanted to view the contents of a collection, click on the title and that will take you back to PubMed. In the Settings/Sharing column, collections are set to either private or public. So private means that only you can access them, and if you want to share this collection, we’ll need to switch it to public.

So, to do that click on the year beside the word Private. And on this page under collection sharing select Public and click save. So here is my collection of results. And you can copy and paste the URL from the top of the page and send that to whoever you want to share your collection with. I'm going to paste into the chat this link to the collection so that you can see what it looks like when someone shares a collection with you.

So those are a few ways that you can save or share your results and, besides results, you can also download your search history and details. Which includes the search terms that you used, any filters that you applied and the number of results retrieved. You can do that from the PubMed Advanced Search screen so I’m going to go there now.

I'll go back to PubMed and into the Advanced Search screen. And then scroll down to the History and Search Details box. Now as you probably are aware, if you click on the carrot symbol under details for a search, you’ll see how PubMed translated your search terms and any filters that you have applied. So, all of this information will be included when you download. And to download a csv file with this information I will just click download from above the results column. And there it is.

So those are a couple of options for saving and sharing results and then your search history and details. I'm going to switch back to my slides. And to wrap up this demo if you would like to learn more about PubMed, we offer live and on demand trainings.

We're going to put a link to the training schedule in the chat and coming up in April is our “How PubMed Works” series if you would like to go in depth into PubMed’s contents, automatic term mapping and MeSH. We have an on-line training page with trainings in a variety of formats including interactive ones and previously recorded classes.

Kate will also put a link to that in the chat and to learn about upcoming trainings and any PubMed updates sign up for the NLM Technical Bulletin which is where we announce all of those and Kate will also put a link to those in the chat. And I’ll hand it back over to Mike for questions.

>> Awesome. Thank you so much, Catherine. We've got a bunch of great questions. Somebody is asking if you could show what the csv version of the search looks like?

>> yes. I can absolutely show that. One moment while I open that up and I will share my screen. Do you see an excel -- spreadsheet?

>> yes, we do.

>> okay. Great. I've only got one search in there so if you're doing this and you have multiple searches, they will each be in their own row, but this shows you the query so, it’s giving me what I searched and then the filters. In their own column so, whatever filters applied and then the search details are where that translation is. And it will also let me know how many results I got. So again, if you have multiple searches, they would each be in their own row.

>> And, Catherine, can you also show viewing search results not the history but viewing the actual results and saving the file as a csv.

>> yes. I can do that. I'm going to stop sharing for a moment. And grab those. Here we go. I'll reshare my screen. So now you should be able to see a different excel and this is if you save your results as a csv file. You can see across the top row that you can sort by title, author, journal, whatever you need, and this is the information that it brings along with it.

>> great. Can you make those columns a little bit wider so we can see what is in there?

>> absolutely.

>> thanks.

>> I’ll run out of screen.

>> I wanted to make sure -- that people could see what is in there and what that looks like. Of course, if you're interested you can go to PubMed with any search results and open it up in excel. All right. We've got a bunch of great questions coming in. So, it will take us a minute or two to sort through some of the questions. But I’m going to start with a question from Amanda.

Will recent activity in myNCBI be improved?

This feature has worked sporadically. I'm going to go to Kathi Canese with that question.

>> Recent activity is built on the same system that the other Entrez databases and legacy PubMed was built on. But when we transformed the new PubMed to the cloud and to our updated technology it is running on a different platform and unfortunately, we cannot offer myNCBI recent activity at this time. We have a Technical Bulletin article that describes what that new technology looks like if you’re interested maybe somebody can pop it in the chat or you can find it in the TB.

>> we'll get that -- and post it in the chat. Thanks, Kathi. This next question is from Amy.

When you share a collection using a public link, is the search strategy also shared or just the list of results?

I'm going to go to Amanda Sawyer.

>> sure. Thanks for that question, Amy and the answer is no it will not share the search and the reason is because collections are only saving the items that you send. You could add citations from multiple searches to that same collection so, it's not going to share the searches just the citations.

>> Awesome. Thank you so much. The next one I’m going to go to -- whose question? I just lost it. There it is. Erin's question.

What do the initials NIM and IM mean? This is for folks who are using the MeSH browser to review the MeSH vocabulary.

We're going to go to Katherine Majewski.

>> Sure. The NIM and IM terminology that you’ll find in the MeSH browser is intended for indexers to use. An IM term means that it’s terminology that an indexer could assign as a major point of the article. Like organs or diseases or organisms or chemicals or therapies. These types of concepts can be assigned as major topics but there are some kinds of terms that cannot be assigned for example techniques, animals that are studied experimentally and some other terms. So that is what that means when you come across it in the Browser. Thanks for the question.

>> And thank you for the answer. There is a couple of questions here that I’m going to sort of group together. Actually, I am going to hit this one first. A part of this category sort of. Some questions are coming so fast and furious I’m sometimes scrolling past them. There is a question from Diane.

Are there any plans to export affiliation into one of the formats?

Affiliation and all of the other fields are included in both the PubMed file format and the .nbib file. That information is included and that should include the affiliation. And our panelist is pointing out that affiliations are also included in the abstract text format. So those are formats that do include that author affiliation information.

Let's see. What else we can clear up for people today? If you. Hmmm ... Trying to find the right person for the right question. This one is from Deborah.

Are there plans to expand the translation of search terms that are automatically mapped to both the singular and plural terms? For example, misdiagnosis and misdiagnoses.

>> If the terms are MeSH terms those can be added relatively easily. On the NLM MeSH home page there is a place where you can click on suggestions and suggest we would like the plural to match the singular or singular to match the plural. The MeSH Section creates a MeSH mapping file and we import that into the PubMed translation tables. If it's not a MeSH term that is more complicated. We curate those data, and we have our computational biology branch that is building additional taxonomy outside of the MeSH environment. So if you have examples, if you just write to the PubMed Help desk then we can share those with the team and hopefully expand that a little bit more.

>> Kathi. That is great. And that brings up something else. A couple of folks who are asking questions about feature suggestions or upcoming possible features or possible changes. And I want to reiterate what Kathi said is if there are things that are generally not working the way that you would expect, or feature suggestions, the best way to get those heard is to send them through NLM customer service. At the bottom of every PubMed page there is a Help link.

You can send a message through to our help desk and those feature suggestions and things, and when you submit them that way, not only does the PubMed team see those things -- but those are things that are quantifiable. They can take a data driven approach to how many people are requesting what sort of things and that can actually help inform future development.

Okay. So, changing gears a little bit. I saw that Joe had a question about

the different dates in PubMed and what they mean.

Kate, I know that you've taught classes on this before.

>> Sure. And actually, I’m going to take advantage of this opportunity if I can to share my screen. Because I would love to show you how to find details about the different searchable fields in PubMed. Okay. So, you're seeing PubMed on your screen right now. And if you scroll down to the menu on the left where you see the FAQs and User guide. On the page navigation on the right of the user guide you will see the -- Appendices. And one of those appendices is the Search Field descriptions and tags.

I was looking in the user guide on the right-hand side of your page navigation under Appendices -- if you ever want to know more about any particular searchable field in PubMed you can come to this table and learn what is the data in the field. So rather than go through every different date field for you I will point you to this resource and you can find the descriptions of the different date fields. The entry date. The create date. The date that goes into PubMed. So, take a look and if you have questions, please write to us or if you're still on-line send us a chat today. Thanks.

>> Thank you, Kate. And our next question -- let’s see -- yeah. Okay. This is a question from I lost it again. Sorry. From Alissa.

Dealing with records that have exceedingly long abstracts that scroll on for ages. She seems to get several in any search, and it can be difficult to move through them. Amanda, do you want to talk about ways to accommodate this. Maybe do a screen share.

>> Sure. Let me screen share and find the right browser. Perfect. If you're using this display option one of the ways to clear up your search to be able to scroll a little bit better – it’s a recent feature. In your display options you can select or deselect the show snippets option and this is something that we found really clears things up for people who are having scrolling issues. Alissa does that answer the question that you had -- or are you seeing a different issue?

>> She has to read abstracts. Unfortunately, there is a limited amount that we can do about that. The abstracts are the length that they are. We don't want to trim them or shorten them because folks who need to see that information or have access to that information may require that. There may be another answer. I might circle back to your question again but first I want to go to a question I think this is another question from Hillary about public collections.

>> And I think Kathi might be able to answer this one.

If you send the public collection via the link to another person, can they add that collection to their own, so they see updates to it and don't have to bookmark the URL? You have something that you can say to that?

>> Actually, they can continue to access that public collection whenever and if the owner of that collection adds the additional citations and of course they would see those additional citations. If, however they want to take that collection and make it their own they can do that because once they click on the collection all of the citations will display and then they can just create a collection using that citation list.

You cannot merge public collections. You cannot group edit them and I think that is what you're after. That is not possible unless you create a shared myNCBI account and you share the password and the username with everybody and then you can do it that way. But they are sort of siloed into specific myNCBI accounts. So, and also and I hope that answers the question.

Let me circle back to Mike’s explanation about the abstracts. You're correct. There is nothing really that we can do. The publishers submit this data and we do not want to touch the data. We're not the publishers. We're the folks that index it and present it to our users. I know sometimes it's a pain to have to deal with those long abstracts but that is where we are now. Thank you.

>> Thanks for that and I will say this is a tough issue. That because we do -- we do -- we’ve often received feedback that not having all of the author information and affiliation displayed on that brief view is a huge problem for other people. So, it's a difficult thing to balance.

We do encourage you and we appreciate the feedback and send the feedback including examples of how you use PubMed and why things are specifically not working in a way that is working for you.

Obviously, we can't always make everyone 100% happy all the time but by getting a better sense of the ways that people are using PubMed and how the features and the way PubMed works is impacting that, that can definitely influence development down the road and even in the short-term.

There are, as I think Catherine pointed out, the bar of the saving citations and sending citations controls that previously appeared at the top of the results list and now sort of floats with you as you scroll down the results list that was a feature that was introduced in response to user feedback.

So, we're listening. Unfortunately, we cannot -- we often get diametrically opposed feedback. So, there is a balancing act that we have to do and again we encourage you to submit your feedback via the customer service link. All right. If y'all will stand by for a moment I’m going to take a look to see what questions I’ve missed.

>> Here is a question that Alex submitted.

Is there a way to make an alert to send to an account other than that of its creator? That automatically e-mails to somebody other than the person who set it up.

Let's see. Anybody want to take a crack at that, Kate is that something that you can address?

>> I am happy to grab it. Yeah. For myNCBI accounts, those must include a single e-mail address. And because we want to make sure the system is relatively secure, and people are not spamming other people—, you can only create myNCBI searches and e-mail alerts for the e-mail that is registered for that account. If you want to go through the painful process of creating additional myNCBI accounts for your users, you could do that. And then have those e-mailed or go to the e-mail address associated with those accounts.

>>Excellent. Thank you for that, Kathi. I have been workshopping another answer to Alyssa’s question -- I don't have a perfect solution. But if the searches that you’re doing specifically if you’re seeing conference abstracts come up and you don't ever care about conference abstracts you could add to your search -- sort of NOT out the publication type for conference abstracts. That would exclude them from your search results -- or meeting abstract might be the right one. I have to double check.

Again, like I’m -- trying to do this off the top of my head. There are as you said certain types of articles that have lots of authors and lots of abstracts. I believe there is a way that you would be able to set up a search strategy to not include those but that would again require that you were not interested in seeing those.

So again, something that I would to have think about more in order to make sure you’re getting all of the results that you’re looking for. So, and I don't know that I can do that quite on the fly. Maybe I’ll think about it a little bit more while Catherine answers this question.

>> Another question about Collections. I can't find who asked it.

If you put a citation that is ahead of print in your Collection file, will it update the citation information automatically once that article goes into print?

Yes. The answer is it will, so the citation will get updated by the publisher with final publication details and that will automatically show up for you in your My NCBI collections and in PubMed.

>> excellent. While again there have been so many great questions in the chat and I want to make sure I don’t miss anything. So, while I’m looking through that I’m going to pose a more general question to our panelists. And that question is what is the question that you most commonly get asked about PubMed. What is your number one question that you hear and also if you don’t mind what is the answer to that question? I'm going to start with Kate.

>> As a PubMed trainer most often I get the question of why am I receiving results in PubMed for records that don’t include my search terms? So, I think that if you’re looking for your search details that it might be confusing. So, I’ll show a quick example. I hit the wrong button. Let me try that again.

So, I think that introductory PubMed users are not familiar with things like term mapping. PubMed is doing a lot of things behind the scenes. Attend our PubMed training or take some of our tutorials but when you're in PubMed, you can click on advanced and scroll down to your search details which appear at the bottom here and you can see all of the mapping for your individual terms including the translations to MeSH.

So, the most common question I get about PubMed is where are these other terms coming from? You can always check your details.

>> I hear that one a lot too especially in training. Kathi, the same question.

>> I think the vast majority of questions that we are getting mostly from authors that are concerned that their name is incorrectly entered into PubMed. We get authors, librarians on behalf of authors. We get publishers and what we recommend that those users do is reach out directly to the publisher and the journal because as I mentioned before we don’t have control of those data. Well, we have control, but it is the onus of making those corrections is on the publisher’s side.

They submit those data to us directly. We deposit it in PubMed and provide both a UI interface to the publisher as well as a file sharing system for them to make updates and corrections. So, we reach out to the -- when users send us, or authors send us those concerns we do reach out to the publisher directly, but we encourage the author of course to contact them directly as well. So, in a way it's quite nice now that publishers have the ability to make their own corrections because PubMed gets better and better every day and cleaner because it's constantly being updated and corrected.

>> excellent. Excellent. All right. I think we've mostly caught up with the questions in the chat but if I missed your question, please submit it one last time. We have a few more minutes.

Is there a limit to the number of collections one can have in My NCBI?

>> sure. Thanks for the question and no. There is no limit on the number of collections you can have.

>> awesome. Thank you so much. Let me take one last look around for questions. Joe has a question. Kate, you want to take this one.

The difference between MEDLINE and PubMed.

>> Sure. This is another common question. MEDLINE is a database of a set of journals that are selected through a special advisory committee to NLM. And so, MEDLINE is a subset of the PubMed records. It is consists of the vast majority of PubMed records but not all of them.

There are other records that we get from PubMed Central which has a different parallel acceptance criteria for those journals and some other sources as well. In fact, we have a video that talks about all of the different ways that records get into PubMed so I’m going to in the last minute -- I’m going to find a link to that video for you. Thanks.

>> People's questions have come in fast and furious here. We have basically a minute left. I'm not sure we're going to get to everyone, but Allison is asking

Is there a way to have links to repository versions of publications in PubMed records.Kathi.

>> we have them already. And if you go to the LinkOut documentation which is a little sketchy, we're planning on updating that soon. You can find a link to all of the LinkOut institutional repositories that we now currently are linked to.

>>I see two more questions. They both have to do with errors. Other than -- not errors necessarily about citations. The question is –

If the citation is correct in PubMed but the author's name is wrong in the APA and NLM format

I think that is a great opportunity to go to the help link at the bottom of the page and send that into customer service. Similar question about errors in MeSH indexing. -- send them into customer service.

We are 100% out of time. Catherine, can you bring up your slides one last time. All right. There are plenty of ways to get help about PubMed. Including our upcoming NLM trainings. I believe there are links already to that in the chat. The NLM Technical Bulletin is the best way to find out new information about upcoming events and trainings and future NLM office hours. We will be doing these again. They will feature other products as well but stay tuned.

And as we've said many times NLM Support Center is the best way to get answers even between these types of sessions. We love being able to answer your questions but there are faster ways sometimes to get answers than waiting a month or two months for the next opportunity. And that opportunity is the NLM Support Center. Catherine, can you advance one please.

When you leave this session today you will get a pop-up for an evaluation form. We encourage you to fill out the evaluation form. This is a new thing that we’re doing. We're not sure how it needs to change and grow. What we could be doing better. Please fill out that evaluation if you want CE credit. Follow the instructions on the screen.

Thank you all for coming. Thank you to my fantastic panelists and hopefully we will see you all soon.

For more information PubMed Home.