Talk:PubMed

From Wikipedia, the free encyclopedia
Jump to: navigation, search
          This article is of interest to the following WikiProjects:
WikiProject Bibliographies (Rated C-class, Top-importance)
WikiProject icon This article is within the scope of WikiProject Bibliographies, a collaborative effort to improve the coverage of Bibliographies on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the quality scale.
 Top  This article has been rated as Top-importance on the importance scale.
 
WikiProject Academic Journals (Rated C-class)
WikiProject icon This article is within the scope of WikiProject Academic Journals, a collaborative effort to improve the coverage of Academic Journals on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 
Note icon
This article has been marked as needing an infobox.
See WikiProject Academic Journals' writing guide for tips on how to improve this article.
WikiProject Medicine (Rated C-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Medicine, which recommends that this article follow the Manual of Style for medicine-related articles and use high-quality medical sources. Please visit the project page for details or ask questions at Wikipedia talk:WikiProject Medicine.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.
 
WikiProject Databases / Computer science  (Rated C-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Databases, a collaborative effort to improve the coverage of database related articles on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.
Taskforce icon
This article is supported by WikiProject Computer science (marked as Mid-importance).
 
WikiProject National Institutes of Health
WikiProject icon This article is within the scope of WikiProject National Institutes of Health, a collaborative effort to improve the coverage of the National Institutes of Health on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
 
WikiProject United States / National Institutes of Health (Rated C-class, Low-importance)
WikiProject icon This article is within the scope of WikiProject United States, a collaborative effort to improve the coverage of topics relating to the United States of America on Wikipedia. If you would like to participate, please visit the project page, where you can join the ongoing discussions.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Low  This article has been rated as Low-importance on the project's importance scale.
Taskforce icon
This article is supported by WikiProject National Institutes of Health.
 

Tutorial[edit]

I'm planning to add direction on "how to use" pubmed within the next month. Does anyone have suggestions or comment? Alot of the work will be based on the linke below, or perhaps I should just link to it. My idea is to have people search this database to be able to cite good then in the Wiki.

http://www.ncbi.nlm.nih.gov/books/bv.fcgi?rid=helppubmed.section.pubmedhelp.Understanding_Your_S

meatclerk 18:51, 15 May 2006 (UTC)

I'd be inclined not to. The Pubmed tutorial on their site is quite long and complex. It's not quite the role of an encyclopedia to act as a manual. But I'm open to suggestions. Mccready 02:06, 16 May 2006 (UTC)

Actually, I plan to make a condensed two (2) page printable version. The basics, type in the searh box, your results mean this, then an explaination of opening an account, no more.

The biggest problem with their tutorial is that it reads like documentation, not a real tutorial. A tutorial should be enough to get your feet wet, then point (in a bullet format or other form) where to find other resources or instructions. That's about it.

meatclerk 07:55, 16 May 2006 (UTC)

Ok, let's see what you come up with. have fun :-) Mccready 11:44, 16 May 2006 (UTC)

This note is posted with some disappointment. For now, I am putting off indefinitely the authoring of the tutorial I suggested. meatclerk 16:23, 17 September 2006 (UTC)

subjects covered[edit]

I added a brief statement; The paragraph for nursing was disproportional to the other disciplines. DGG 17:35, 4 February 2007 (UTC)

I added recently the following entry to a very useful site (PubMedReader) but somebody deleted it. Why? It's free and open to anybody. There is not even ADVERTISMENT? Please leave this entry there. Thanks!

  • PubMed Reader - A free web-based research program for displaying customized PubMed search results

Question[edit]

I have (in vain) tried to find out, what "PubMed" stands for, though searching on the homepage on pubmed a long time and googling as well. I think this should, no: must be explained on Wiki. Pub: Public? Publications? Med: Medicine? Medical? Medline? "Public Medline" as distinct from versions of Medline intended for professional use by experts. Such versions are still around, developed and marketed by Ovid, SilverPlatter, WebofScience, and dozens of others. Professionals still often prefer them--they are smoother, though they do no more in essence. DGG 08:58, 10 April 2007 (UTC)

A comprehensive answer to this question is given by NLM here: http://www.nlm.nih.gov/services/pmname.html

Question: What does the name PubMed stand for?


Answer: - When we came up with the name PubMed, we were not trying to use an acronym or an abbreviation. However, we can say that: - the 'Med' part of the name refers to the MEDLINE database, which is what PubMed searches - the 'Pub' part of the name can be thought of as either public (since PubMed is the free version of MEDLINE) or as publisher (since PubMed includes links to publisher web sites). —Preceding unsigned comment added by 91.104.23.41 (talk) 08:22, 24 September 2007 (UTC)

Biowizard[edit]

Its not primarily a front end to PubMed, though it does offer that function. its a "journal club" type of site where people nominate PubMed articles to discuss. I see no indication its affiliated with PubMed. There are some similar sites, run by BMC and others. It looks interesting, and if anyone has written an article or two about it, it would perhaps be appropriate for an article of its own.DGG 08:58, 10 April 2007 (UTC)

Re adding content without prior discussion is not good practice, and if continued s usually considered vandalism. DGG 04:25, 12 April 2007 (UTC)

Genomediff[edit]

A very important program, but it should go probably under Entrez--its not really part of the core Medline/PubMed. And we seem to badly need a general article on genome databases, though we have articles on individual ones. --perhaps you could write one? DGG 05:43, 6 May 2007 (UTC)

Language coverage[edit]

What is the language of the indexed articles? Only English? --84.20.17.84 07:19, 24 August 2007 (UTC)

No, all languages can be indexed, including non-western languages (mainly Chinese and Japanese). These are indexed with an English title in square brackets, and a notice of the language of the article. Often an English abstract is also provided. A random example 217.122.83.79 16:45, 11 September 2007 (UTC)

how to search pubmed[edit]

It is not usual to includes this sort of detail in WP articles. I know its valuable. I recognize that the advice is good advice. Please try to fid some way of doing it without giving sample question and hits, talking rather about the structure of the database and the way it parses search queries. DGG (talk) 07:20, 14 October 2007 (UTC)

Telegram-Syle Searching[edit]

It is a pity that this section has now been shortened and that some actual 'telegram' search examples have been deleted. It is constantly and everywhere suggested that PubMed is difficult to use (that's why there are so many 'alternative' interfaces). I believe (and could provide the references) that PubMed is easy to use, just like Google. I dislike all those artefacts (field searching, e.g. [ip]). Most clinicians use the 'telegram', probabilistic style, and not the protocol-driven complex search strategies. What a pity that the few telegram-style search examples have gone - I put them there, so I am of course biased (?). I accept that a full tutorial has no place in a Wikipdia article of this kind, but a few actual examples of how CLINICIANS use PubMed??? Reinhard Wentz, London

I've remove the section per WP:NOTHOWTO. --Ronz (talk) 15:17, 23 October 2008 (UTC)
Giving search examples does not constitute a Tutorial OR 'How to'. They just show an easy way of how PubMed is being used (by most people). BTW: What are the complex Tag searches doing on this page? A list of tags would be enough. The Boolean search examples are of very doubtful value. Why not delete them, and just leave the notes on Boolean searching (you deleted my amendments. Why? They are factual and not tutorial) I put them back.
I hope my 'additional functionality' bits (Related Articles, Clinical Queries, etc.) are acceptable and will stay. I left the search examples out.
Reinhard Wentz sleuthmedical@yahoo.com 23.10.08
Thanks for responding. I'm not sure what to do with any of it. None of "Searching PubMed" section seems to be appropriate for an encyclopedia article. --Ronz (talk) 17:28, 23 October 2008 (UTC)
as a purely practical problem, the Pubmed search page is notably obscure and difficult to use properly, and the help is fairly well hidden. A list of searchable fields is a reasonable part of the description of a database. As for Boolean, PubMed does not use pure Boolean searching, but an approximation to it--see their help--so a description of which of the normal search functions work and how they work is appropriate also. From experience writing these things, generally examples are the clearest way. Perhaps we can find a suitable help guide that has them for us to link to. DGG (talk) 04:03, 24 October 2008 (UTC)
Thank you DGG for these comments. I agree that the PubMed search page is complex, but why obscure? Of course it is difficult to use it properly! Finding relevant, current and valid medical information is a difficult process requiring training and experience, as mentioned in the current version of WP PubMed. My point is that PubMed also allows for simple, focused question to be answered quickly if a simple search 'telegram' formulation is entered in the search window (ignoring all the complexity of a comprehensive, exhaustive search). (Many ‘alternative’ Medline interfaces reflect this and offer a simple search window, without PubMed optional search sophistication).
There is a further important point (absent from most discussions of How To Search Pubmed): clinicians usually do not require a complex, exhaustive search: they already have a good (probalistic) idea of possible diagnoses and treatment options and just want to do a quick search to increase the probability that they are on the ‘right track’. A excellent example of this ‘Bayesian’ approach to searching (The priors? 'Given that I am a clinician...) is provided in the paper by Glasziou (PMID: 17884906) which I used in my original set of ‘telegram search formulations’: He did a ‘quick search’, found some eight papers and used inter alia the abstract from only one of those hits to confirm his treatment choice. Brilliant! See his defence in the Rapid Response section to his article (http://www.bmj.com/cgi/eletters/335/7620/612#177973). Sorry for being verbose. Reinhard Wentz
this doesn't refer to some of the alternate interfaces. Most of them are reasonably good. The very best one, though, was in my opinion a fill-in-the-boxes interface called "Internet Grateful Med" which NLM withdrew a number of years ago--in my cynical opinion because it was so good it competed favorably with the commericial interfaces. DGG (talk) 04:43, 25 October 2008 (UTC)
Your comment re 'Grateful Med' is indeed slightly cynical. A very reasonable explanation for the NLM's decision to drop Grateful Med is given here: http://www.nlm.nih.gov/services/igm.html In addition: PubMed has now an 'Advanced Search Beta' interface which allows fill-in-the-boxes searching. Reinhard Wentz

Citation Searching on PubMed[edit]

What's this section doing in a Wikipedia PubMed article? PubMed is not a citation index. This section should be deleted. Lack of citation data is NOT a limitation of PubMed! A motorbike hasn't got four wheels, so what? Google Scholar doesn't map to subject headings, so what? Reinhard Wentz, London

agreed. But a statement of what the "related articles" actually are is appropriate. DGG (talk) 04:03, 24 October 2008 (UTC)
Agreed. But I hesitated to give examples lest this would constitute a tutorial. A simple reference to how the 'see related article' algorithm works (from Pubmed Help) would perhaps suffice. Reinhard Wentz

I am not too good with editing a WP article. I would have further points but perhaps we can continue this thread via e-mail? Reinhard Wentz, sleuthmedical@yahoo.com you ! americans and europeans what do you think about asian medical experienses and progresses? —Preceding unsigned comment added by 213.52.6.73 (talk) 19:06, 5 May 2009 (UTC)

Criticism[edit]

The recently added critcism

'The use of tags and boolean operators makes this search engine far less powerful than mainstream equivalents. This is mostly due to the fact that there are so many operators that users cannot remember them. The PubMed search engine is effective, but for most scientists it is a poorly constructed interface more akin to a computer programmer than a day-to-day user'

seems unjustified. As mentioned elsewhere in the article proper, PubMed can be searched without using tags or Boolean operators. The criticism is not supported by a reference and I am minded to delete it. Reinhard Wentz 22.07.09 —Preceding unsigned comment added by 91.106.26.96 (talk) 09:50, 22 July 2009 (UTC)

New PubMed Interface October 2009[edit]

I added two notes on the new PubMed interface, launched in October 2009, one under 'Quick Searches', one under 'Alternative Interfaces'. Reinhard Wentz 07.11.09 —Preceding unsigned comment added by 91.104.15.218 (talk) 08:47, 7 November 2009 (UTC)

Using PubMed is easy[edit]

It is a pity that some well-meaning contributors make using PubMed appear difficult. For instance, knowing (and using) field qualifiers is usually not necessary; one does not need to know that e.g. 'PNAS' is an abbreviated journal title, for which the field qualifier would be [TA](in square brackets!). A search in the form ' pnas drexler ke 1981' correctly identies the single matching paper in the journal PNAS by KE Drexler, published in 1981. —Preceding unsigned comment added by Sleuth21 (talkcontribs) 11:28, 8 July 2010 (UTC)

I was wrong in not noticing that PubMed makes guesses that can often be correct. However, giving an example showing the correct invocation of defaults was not a good way to explain how PubMed tags and operators work together. I have expanded the example to make more clear how tags and operators work together, as well as illustrate correct and incorrect invocation of PubMed defaults. --Ben Best (talk) 12:54, 15 July 2010 (UTC)
I can see your points, BenBest, but PubMed doesn't ‘make guesses’ when 'assuming' e.g. tags but uses a clear probabilistic algorithm to do so, just as it does for ‘Textword to Medical Subject Heading (MeSH)’ mapping. Your example ‘78 [vol] 5275 [page] 1981 [dp]’ is a very special case (a search using numerical values only), useful when verifying a reference, but even then the PubMed's ‘Citation Matcher’ should be used. Most searches on PubMed by clinicians and members of the public involve 'facetted' subject searches, author names, perhaps a year of publication etc. where PubMed’s mapping algorithms work very well e.g. ‘pnas drexler ke 1981’ as does (for the same unique paper) ‘5275 pnas drexler’.
Further, a more realistic example of how Boolean operators are processed would be e.g. ‘Spleen AND Cats OR dogs’ (wrong) and ‘Spleen AND (Cats OR dogs)’ (correct) with the invocation 'Use brackets when mixing Boolean operators’. That works in all cases – ‘All Boolean operators are processed in a left-to-right sequence’ is not very helpful / too technical.
Also (a minor point): does one use [page] or [pg] for the pagination tag? Both forms appear in the article and both a valid.
I left your amendments as they are... —Preceding unsigned comment added by Sleuth21 (talkcontribs) 12:45, 25 July 2010 (UTC)
BenBest: on long reflection I decided to delete your redundant complicated amendments of July 15. You re-introduced them. You obviously have never tried to explain PubMed (or another complex database) to a non-expert. Why make life so complicated when a simple statement covering the vast majority of search formulations would do? Verifying a reference is NOT a standard question on PubMed, and in any case its 'Citation Matcher' should be used for it. See also my earlier comments above.
Using PubMed is NOT as complicated as you insist on suggesting, BenBest

Merge proposal[edit]

I can't think of any good reason for the stub PMID to be in a separate article. There just isn't very much to say about it. WhatamIdoing (talk) 00:09, 29 August 2010 (UTC)

PUBMED Science Citation Index relation[edit]

Does anybody know the relationship between PUBMED and Science Citation Index? Does PUBMED contain all SCI publications? Does it contain all SCI-Expanded publications? Does it contain more than SCI-E? —Preceding unsigned comment added by 78.166.133.235 (talk) 13:43, 16 January 2011 (UTC)

  • There is no formal (or informal, as far as I know) relation between the two. PubMed is non-profit, operated by the US National Library of Medicine, SCI is operated by Thompson Reuters, an international company. They determine their coverage independently, with SCI being much broader than PubMed (which only covers medicine-related topics/journals). Hope this helps. --Crusio (talk) 14:18, 16 January 2011 (UTC)

PubMed radical edit (03.03.11)[edit]

This WP article needed editing but I doubt whether Crusio's radical edit, deletions, and amputations are justified. Some should have been discussed here. The article is now smarter looking but poorer in content. Healthy tissue was cut off. Pity. —Preceding unsigned comment added by 91.106.49.201 (talk) 21:08, 7 March 2011 (UTC)

Yes. That was inappropriate of User:Crusio (to have done so without discussion first). I have restored most of the material. --Thorwald (talk) 09:05, 17 March 2011 (UTC)
  • All that stuff flies in the face of WP:NOT (and also WP:OR for that matter) and should be deleted again. Removing material that goes against policy is not inappropriate, restoring is is. --Crusio (talk) 09:12, 17 March 2011 (UTC)
    • That would be true for most articles. However, this article is about a service and that service can not be understood without examples of how to use it (for the average layperson). The WP:NOT are guidelines, not rules (set in stone). This article is an example where those guidelines can be skirted. --Thorwald (talk) 09:20, 17 March 2011 (UTC)
      • You're wrong. The material you restored is against policy in multiple ways. You restored unneeded/irrelevant material, inappropriate in-text external links, etc, without giving any valid reason for that, other than "I don't like it". --Crusio (talk) 09:41, 17 March 2011 (UTC)
        • The material I restored was necessary to provide context to the article and for the user. I am a research scientist, so I know how PubMed works (I use it pretty much every day). However, for the average reader of this article, the purpose of this article would remain unclear if they did not see how to use PubMed and why this article is on Wikipedia in the first place. My restorations/revert have nothing to do with "not liking it". If there are "inappropriate" external links, we can discuss these (please provide specific examples of these). However the majority of what I restored was not "unneeded/irrelevant" material. Please assume good faith first before resorting to lesser assumptions. Thank you! --Thorwald (talk) 09:53, 17 March 2011 (UTC)
          • How often do I have to repeat this: Wikipedia is not a how-to manual. The stuff is unencyclopedic, uses formatting that does not adhere to WP:MOS, includes material on subjects other than this article (which have their own articles), uses in-text external links (WP:MOS), etc. In addition, the PubMed website has, of course extensive help pages itself. Those pages are where that material belongs, not here on WP (and in addition, these help pages are the first link appearing on PubMed's home page). --Crusio (talk) 10:21, 17 March 2011 (UTC)
            • Okay. I agree that most of the HOWTO material can go (since, as you correctly pointed out, the PubMed site has an extensive page on just this). However, I do not agree with everything you removed and we need to discuss those as well. --Thorwald (talk) 10:36, 17 March 2011 (UTC)
              • Well, go ahead and re-delete that stuff. And then tell me what you think should remain and we'll talk about that, too. --Crusio (talk) 11:18, 17 March 2011 (UTC)
                • Done. I have removed the HOWTO/tutorial material. --Thorwald (talk) 12:10, 17 March 2011 (UTC)
  • (unindent). Now about the rest. Lots of edits that I made have been reverted, too. This is a database, entries are called "records", not "citations", to mention just one detail. External links should not be used in the text of an article [[WP/MOS]) is another problem. There is a large discussion of the Consolidated Appropriations Act of 2007 which concerns PubMed Central, not PubMed and thus does not belong here. Wikilinks to PubMed itself have been re-introduced, etc etc etc. All these issues need to be (re-)addressed. --Crusio (talk) 13:22, 17 March 2011 (UTC)
    • 'Now about the rest', 'All these issues need to be (re-) addressed', 'How often do I have to repeat this?': this is all quite intimidating. The issue for instance of tutorial elements in the PubMed article has been discussed more than two years ago and some helpful tutorial-type information was allowed to remain and even survived an 'under dispute' flag. An enterprise like WP of course needs rules, but they have to be interpreted in the (great) spirit of the project. Mere references to 'WP:MOS' etc. don't help an infrequent WP contributor. Assumption of 'good intent' or 'good faith' should, at least sometimes, be allowed to trump strict interpretation of guidelines. Anyway, I prefer the PubMed article as it was before March 2011 ('telegram-style search formulations': Great! 'Google-like searching': Great! 'Usually no tags / field names required': I do like it!) and have a permalink to it: I will refer PubMed punters to it if required. —Preceding unsigned comment added by 91.104.11.200 (talk) 22:51, 21 March 2011 (UTC)
  • That is perfectly fine of course (the permalink). However, "assume good faith" means that one should assume that an editors' actions are done in good faith, not that incorrect edits should be accepted because they were made with good intentions. And even a newbie can click on a link if a reference is given to WP:MOS, and see what that (and linked sub-guidelines) is about. And, of curse, if you think my edits are wrong, feel free to start an RFC (request for comment). As Thorwald is not responding, I will address the remaining issues in this article as soon as I have a moment. --Crusio (talk) 02:58, 22 March 2011 (UTC)

PubMed stats[edit]

The updated figures added by me in the last paragraph under 'Contents' are not taken from the item mentioned in reference 4, but rather from a recent search on PubMed in the form 'all[sb]', which can then be limited to 'abstracts' and 'free full-text' respectively. I have been told that a statement to that effect wouldn't constitute a proper citation (strictly speaking true I suppose), so I quite expect the old, outdated figures to be re-instated, as has happened before. —Preceding unsigned comment added by 91.104.16.243 (talk) 19:47, 24 March 2011 (UTC)

I find that a strange thing to remove. In fact, User:Crusio commented out a perfectly reasonable, reliable, and continually updated way to find the current (as in, daily updated) number of records in PubMed. This is what Crusio commented out "(Note: To see the current size of the database simply type "1800:2100[dp]" into the search bar and click "search".)". Crusio claimed, "commenting out non-encyclopedic remark; cannot be used to update the article (would be unsourced OR)". I have two complaints with Crusio's reasons: (1) How on earth is this "unsourced OR"? It is simply a statistical fact straight from the official maintainer of said database. One can not get a more direct and accurate count. Also, there is not "OR" involved; no "research" of any kind. It is just a factual way to find the exact number of records in PubMed at the time of the search; and (2) What is "non-encyclopedic" about it? It is in the References section. If a user reads that PubMed "has over 20.7 million records..." and wishes to verify this fact and know how that number was gotten, he/she can simply input that search criteria ("1800:2100[dp]") into PubMed and instantly verify that the statement in this article was true at the time. Unless Crusio (or anyone else) can specifically challenge that the note in question does not belong in the References section of this article, I will uncomment it. --Thorwald (talk) 17:50, 28 March 2011 (UTC)
  • You should read up on the definition of OR (WP:OR). And no encyclopedia will give its readers instructions like "go to this site, enter XYZ and then you can see for yourself". As for sourcing, we need a written source to substantiate statements. A result from a search engine is simply not good enough. I didn't make those WP policies, but I do see (and agree with) the reasons behind them. --Crusio (talk) 18:10, 28 March 2011 (UTC)
  • Thank you, but I have read the definition of OR many times (I have been a Wikipedian for over 6.5 years). Just read the first two paragraphs of the OR and the "Routine calculations" section and you will find that the note in question does not contradict anything. I also think you are applying the letter of the law, but not the spirit of it. Not all articles are equal, nor are the references/sources to back up the statements found within them. The note in question is not the result of a search engine query, it is a direct question put to a database (which PubMed fundamentally is): How many records do you contain? Would you argue with the many other statistics found in articles throughout Wikipedia as to how they back up their statements reporting the facts of their contents (e.g., Protein Data Bank with its source being taken directly from here, which is nothing more than an automatically updated {weekly} query of a database)? Wikipedia is not a normal, book-bound encyclopaedia. You are being too archaic in your enforcement of this concept. Also note that OR is a guideline, not a rule set in stone. Wikipedia is at its heart a source of information for the curious and a way to come to an understanding of what it states and how. You still have not convinced me that the note in question is not a "source" to back up the claim in the statement reporting the facts about the number of records in PubMed (it is the ultimate source of said information; any report would rely on this query to come up with such a statement/answer). --Thorwald (talk) 18:50, 28 March 2011 (UTC)
  • Thank you, Thorwald, for keeping this discussion alive. To take Crusio’s apodictic 'OR' claim to its logical and absurd conclusion: here are a few external written sources (selection) validating your claim that the phrase "1800:2100[dp]" can be used to establish the most up-to-date number of records in PubMed:
1) http://www.pubemd.org/
2) http://chrtowsky.wordpress.com/2010/04/08/pubmed/
3) http://www.thefullwiki.org/PUBMED and another 100+ written sources
My slightly shorter formulation “all[sb]” can be verified for instance here:
4) http://blogs.ubc.ca/dean/2010/04/pubmed-ovidsp-shortcuts/
5) http://www.ncbi.nlm.nih.gov/books/NBK3827/ [PubMed’s own PubMed Help Manual, natch!]
6) http://hlwiki.slais.ubc.ca/index.php/PubMed_and_OvidSP_shortcuts_%26_command_language and another 200+ written sources
I think your helpful remark, Thorwald, should not just be tucked away in a note, but be mentioned again (before it was deleted) in the body of the PubMed WP article. That would be much better than giving references where this recommendation may be buried deep in the text of the document. Keep up the good argument.
Reinhard Wentz, Twickenham, UK —Preceding unsigned comment added by 91.104.19.189 (talk) 07:37, 29 March 2011 (UTC)

Mapping to MeSH[edit]

I have replaced the incorrect, misleading mapping example (incorrect: 'writers cramp' doesn't map to 'Focal Dystonia', but (misleadingly) to 'Dystonic Disorders' (dystonia can affect not just hand movements but other body parts such as the shoulder); corrected the spelling of MeSH and gave the full form of MeSH. A short descriptionn of the role of MeSH (part of PubMed's controlled vocabulary / thesaurus) will be addded later. —Preceding unsigned comment added by 91.106.42.73 (talk) 07:50, 27 March 2011 (UTC)

Content[edit]

The 'Content' section of this article is much too detailed, confusing, badly referenced, and uses too much jargon. I have drafted an amended version which I will add when I have got a moment. Sleuth21 (talk) 19:32, 29 March 2011 (UTC)sleuth21 Done Sleuth21 (talk) 07:37, 30 March 2011 (UTC)

Comprehensive Search[edit]

Somebody merged the 'Comprehensive search' section with the remark about simple, Google-like search formulations. I will separate them out again and also provide examples to illustrate some of the specific PubMed functionalties. These examples will not constitute a how-to-do manual. Sleuth21 (talk) 07:33, 31 March 2011 (UTC)

New section on 'Standard Searches' added Sleuth21 (talk) 11:01, 12 April 2011 (UTC)

Alternative Interfaces[edit]

The section ‘Alternative Interfaces’ requires a substantial re-think and discussion. PubMed is one of only a handful of professionally curated databases which is available for free and makes its data available under licence, but again for free. It thus has become a playing field for a wide variety of third-party add-on and ‘alternative’ interfaces. There are more than 150 of these, some 25 of which have made an appearance in this section of the WP PubMed article over the years. Some of these ‘alternative’ interfaces compete with PubMed as semi-commercial enterprises; some make justified and/or unjustified (or ludicrous) claims as to offering extra features (in addition to PubMed). So any resource listed here should be carefully chosen.

Then again, one may question whether this section has any place in the WP PubMed article. The WP article on Rolls Royce motorcars doesn’t mention Mercedes or Audi cars. But some useful PubMed add-ons and genuine and free alternatives may be listed here?

I will discuss these issues here and suggest amendments when I have got a moment. Sleuth21 (talk) 12:25, 31 March 2011 (UTC)

Duncan, thank you for starting the revision of the Alternative Interfaces section. Great stuff, but I think you added what I believe are inappropriate cites. Sayers 2010 gives an overview of NCBI Entrez databases, but doesn’t mention any 3rd party PubMed derivates. It is also unusual to give references for derivates, e.g. HubMed, which are mentioned elsewhere in WP with the same reference. I also think the selection of alternatives / derivates is quite arbitrary. It would be much better if the excellent graph of such derivates given in Lu 2011 were used - which I will cite and selectively quote in a future update of this section when I have a moment. Your own paper (Hull 2008) is interesting, but in what way is it relevant to this WP PubMed article? Sleuth21 (talk) 07:26, 27 April 2011 (UTC)
Pubmed.so - a nice litle gadget, but it merely executes a search in the form ‘[subject] filetype:PDF’ on Google / Google Scholar. If the subject is a medical one there are bound to be a few pdf files from PubMed / Medline. If the subject is not medical, no hits are achieved on PubMed, e.g. “Siegfried Line” has no hits in PubMed, on Pubmed.so it gets 400+ hits, none of which is from PubMed / Medline. Pubmed.so has nothing to do with PubMed or MEDLINE; it doesn’t belong to the WP PubMed article. In any case, why are some systems (Pubget or Pubmed.so) so concerned about finding pdf files? What’s wrong with e.g. a file in html format. Am I missing something?

My NCBI[edit]

I intend to add a section on PubMed's 'My NCBI' and some of its salient features when I have a moment. Sleuth21 (talk) 08:33, 1 April 2011 (UTC)

Done. Thank you, Thorwald, for fixing typo and citation Sleuth21 (talk) 23:16, 2 April 2011 (UTC)

PubMed for Handhelds / PICO[edit]

Perhaps the option(s) for PubMed on handhelds (is that term used by digital natives, or only old fogies like me?), especially the elegant NLM-made PICO interface (here http://pubmedhh.nlm.nih.gov/nlmd/pico/piconew.php or here: go.usa.gov/xFn (a charming URL!)) should be mentioned? I may add something if I have got a moment. Sleuth21 (talk) 07:24, 3 April 2011 (UTC) Done Sleuth21 (talk) 10:33, 3 April 2011 (UTC)

askMEDLINE[edit]

... as should askMEDLINE, a free-text, natural language query tool for MEDLINE/PubMed, developed by the NLM http://askmedline.nlm.nih.gov/ask/ask.php, also suitable for handhelds. Sleuth21 (talk) 09:03, 3 April 2011 (UTC) Done Sleuth21 (talk) 12:58, 3 April 2011 (UTC)

LinkOut[edit]

... as should (finally?) LinkOut, a NLM facility to link (and make available full-text) local journal holdings of (mainly) academic institutions, with some 3,200 sites participating (March 2011): http://www.ncbi.nlm.nih.gov/projects/linkout/journals/htmllists.cgi?type_id=6 Sleuth21 (talk) 09:40, 3 April 2011 (UTC) Done Sleuth21 (talk) 12:58, 3 April 2011 (UTC)

Loansome Doc[edit]

... and of course (not finally - there is more!) Loansome Doc. — Preceding unsigned comment added by Sleuth21 (talkcontribs) 12:05, 8 April 2011 (UTC)

Criticisms / Comparisons[edit]

I suggest to add a 'Criticism / Shortcomings of PubMed' and a 'Comparison with other resources' section, both in very general terms. Sleuth21 (talk) 17:26, 27 May 2011 (UTC)

Inclusion/Exclusion criteria?[edit]

I'm wondering why Cosmetic Dermatology (http://www.cosderm.com) isn't in PubMed. --Elvey (talk) 00:02, 20 December 2012 (UTC)

One can't say about a specific journal, but the general rules about which journals are candidates for inclusion in MEDLINE / PubMed are set out here:
http://www.nlm.nih.gov/pubs/factsheets/jsel.html See in particular the sections 'Quality of content' and 'Quality of editorial work' (and related NLM factsheets) Hallodria (talk) 14:15, 21 December 2012 (UTC)
Thanks! (I was going to add a relevant bit to this article, but I see it's already here.)--Elvey (talk) 05:35, 28 December 2012 (UTC)

"PubMed Identifier" capitalization[edit]

The article currently cites PubMed's official FAQ on the meaning of "PubMed Identifier" (which is mentioned more specifically in this section of the FAQ).

The issue is that the article lowercases "identifier" despite it generally appearing capitalized in the FAQ, and another user has used this to justify categorizing the PubMed Identifier redirect as an incorrect capitalization. (This is not meant to be forum shopping; that discussion is about the redirect categorization, while this one is meant to discern the actual correct capitalization of the term.)

Which is right? Should an {{anchor}} template be added to this article to make both capitalizations work as section anchors? --SoledadKabocha (talk) 00:13, 22 November 2014 (UTC)