Jump to content

Wikipedia talk:Identifying reliable sources (medicine)/Archive 12

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 5Archive 10Archive 11Archive 12Archive 13Archive 14Archive 15

Suggestions for MEDRS

Closing discussion initiated by sockpuppet of blocked user Jzyehoshua. Binksternet (talk) 07:31, 24 January 2015 (UTC)
The following discussion has been closed. Please do not modify it.

I think the MEDRS page could use some clarification to avoid confusing new users when the page is cited.

  1. Does the page apply more broadly to all citations of scientific facts, or just medical facts specifically? If so, why?
  2. Pointers on how to recognize systematic reviews vs. primary research. It could be mentioned that systematic/narrative reviews are not research themselves, and thus tend to focus on citing other studies rather than presenting methodology and original tests.
  3. Most importantly, information on how to find sources that meet MEDRS standards. A quick how-to section on how to use the Pubmed Advanced Search might help reduce frustration levels for new users.

I can propose specific edit changes if needed. More clarification and specifics on MEDRS would go a long way toward reducing policy conflicts. --7157.118.25a (talk) 01:40, 17 January 2015 (UTC)

Sure. And we have a box on most talk pages to help people find stuff. Doc James (talk · contribs · email) 02:31, 17 January 2015 (UTC)

Proposed Page Additions

Closing discussion initiated by sockpuppet of blocked user Jzyehoshua. Binksternet (talk) 07:31, 24 January 2015 (UTC)
The following discussion has been closed. Please do not modify it.

I would recommend the addition of a section titled 'Acceptable Vs. Unacceptable Medical Sources' to provide a quick table summarizing good vs. bad sources. While this information generally exists on the MEDRS page already, it is not presented concisely so it can be easily seen. Possible information for the section is as follows:

"Primary sources are articles detailing original research and thus mention methodology used to achieve new discoveries. Secondary sources simply report upon these findings.

Do Use

Acceptable secondary sources that meet MEDRS requirements include:

  • Systematic reviews and Narrative reviews: These are reviews which summarize the history of research on a given medical topic. They will contain links to many different studies. Unlike a research article which focuses on the specifics of a single test, these reviews link to other studies for the purpose of summarizing the history of research on the topic.
  • Medical textbooks: Medical textbooks published by academic publishers are reliable sources. Major academic publishers such as Elsevier, Springer Verlag, Wolters Kluwer, and Informa are preferred.
  • University/Science-Published Health Books: Books published by university presses or the National Academy of Sciences are generally reliable sources.

Don't Use

The following should not be used as standalone sources for medical facts or claims, although they can be used in conjunction with the secondary sources listed above.

  • Research article: These are studies used to arrive at new findings, and should not be used as the main source for a medical claim. This is because research articles primarily outline methods rather than reporting on the results, and are prone to being misinterpreted. Thus reliable commentary on them is needed from a third-party source, i.e. a secondary source.
  • News articles: News articles from the mainstream media are not valid sources for medical claims. News articles are prone to sensationalism. While news articles are secondary sources, they are not scholarly, reliable sources for factual medical information, and thus should not be the sole sources provided for citation of a medical fact/claim.
  • Self-published/Vanity Press health books: Books that are self-published or by the vanity press are unreliable sources of medical information."

I would also recommend revising the section 'Searching for Sources,' a better name for it might be 'How to Find Sources.' I would suggest switching to a bullet point-based system to present the information more simplistically. For example, a section change could read as follows:

"The following are ways to find sources which meet MEDRS standards:

  • PubMed: The NCBI, a U.S. government organization, maintains an excellent advanced search engine for finding medical articles. One should set 'Publication Type' to "review" when running such a search to avoid the chances of encountering primary research articles. Per WP:MEDDATE it is also recommended to set 'Date - Publication' to a date within the past 4 or 5 years.
  • Search Engines: Some search engines have date restriction options. Given WP:MEDDATE it is advisable to search within the past 4 or 5 years. It is advisable to search using specific terms that will appear in professional medical sources, to avoid getting undesired news or blog results."

These section changes could help reduce confusion about how to meet MEDRS guidelines. --7157.118.25a (talk) 04:14, 17 January 2015 (UTC)

Discussion

Closing discussion revolving around suggestions made by sockpuppet of blocked user Jzyehoshua. Binksternet (talk) 07:31, 24 January 2015 (UTC)
The following discussion has been closed. Please do not modify it.
7157.118.25a suggested these changes after our discussion on 7157.118.25a's talk page about what WP:MEDRS covers.
7157.118.25a, I'm not sure about your proposal. It could perhaps help, but it needs some tweaking, given the content that is already in the WP:MEDRS guideline about sourcing and that WP:MEDRS allows WP:Primary sources in a few cases, such as a field that is not well studied. For example, if we add your proposed outline, how would it fit with regard to the Choosing sources section? We've included all of that material in that section after much discussion at this talk page. Flyer22 (talk) 06:02, 17 January 2015 (UTC)
I was thinking of adding the proposed 'Acceptable Vs. Unacceptable Medical Sources' section as the 2nd section on the page, between Definitions and Basic Advice, and leaving the Choosing Sources section as is. The Choosing Sources section does provide good, detailed information; the only problem is that it's very lengthy and not particularly concise. Perhaps the proposed section could contain a Main link to the Choosing Sources section so that users can redirect to the other page section for more detailed information. I am not proposing removal of the Choosing Sources section though, just making a shorter, more readable section containing the most vital Dos and Don'ts higher on the page.
Similarly with the 'Searching for Sources' section, no material need be deleted. Just putting the bullet points at the top of the section would make the key information more visible, and the more detailed information could remain at the bottom. For example, the section could look like this:

The following are ways to find sources which meet MEDRS standards:
  • PubMed: The NCBI, a U.S. government organization, maintains an excellent advanced search engine for finding medical articles. One should set 'Publication Type' to "review" when running such a search to avoid the chances of encountering primary research articles. Per WP:MEDDATE it is also recommended to set 'Date - Publication' to a date within the past 4 or 5 years.
  • Search Engines: Some search engines have date restriction options. Given WP:MEDDATE it is advisable to search within the past 4 or 5 years. It is advisable to search using specific terms that will appear in professional medical sources, to avoid getting undesired news or blog results.
Search engines are commonly used to find biomedical sources. Each engine has quirks, advantages, and disadvantages, and may not return the results that the editor needs unless used carefully. It typically takes experience and practice to recognize when a search has not been effective; even if an editor finds useful sources, they may have missed other sources that would have been more useful or they may generate pages and pages of less-than-useful material. A good strategy for avoiding sole reliance on search engines is to find a few recent high-quality sources and follow their citations to see what the search engine missed. It can also be helpful to perform a plain web search rather than one of scholarly articles only.
PubMed is an excellent starting point for locating peer reviewed medical literature reviews on humans from the last five years . It offers a free search engine for accessing the MEDLINE database of biomedical research articles offered by the National Library of Medicine at the U.S. National Institutes of Health.[1] There are basic and advanced options for searching PubMed.[2] For example, clicking on the "Review" tab will help narrow the search to review articles. The "Limits" tab can further limit the search, for example, to meta-analyses, to freely readable sources, and/or "core clinical journals". Although PubMed is a comprehensive database, many of its indexed journals restrict online access. Another website, PubMed Central, provides free access to full texts. While it is often not the official published version, it is a peer reviewed manuscript that is substantially the same, but lacks minor copy-editing by the publisher.[3]
When looking at an individual abstract on the PubMed website, an editor can click on "Publication Types, MeSH Terms" at the bottom of the page to see how PubMed has classified a document. For example, a page that is tagged as "Comment" or "Letter" is a non-peer reviewed letter to the editor. The classification scheme includes about 70 types of documents.[1] For medical information, the most useful types of articles are typically labeled "Guideline", "Meta-analysis", "Practice guideline", or "Review".

References

  1. ^ Greenhalgh T (1997). "How to read a paper: The Medline database". BMJ. 315 (7101): 180–3. PMC 2127107. PMID 9251552.
  2. ^ "PubMed tutorial: filters". NLM. Retrieved 17 November 2012.
  3. ^ Goodman D, Dowson S, Yaremchuk J (2007). "Open access and accuracy: author-archived manuscripts vs. published articles" (PDF). Learn Publ. 20 (3): 203–15. doi:10.1087/095315107X204012. Retrieved 2008-10-24.{{cite journal}}: CS1 maint: multiple names: authors list (link)

My primary concern is just with presenting more concise, easily read information so people can quickly and readily see what the WP:MEDRS requirements are and how to meet them. If the policy isn't clear, specific, visible, and easily understood, people won't be able to follow it. --7157.118.25a (talk) 20:32, 17 January 2015 (UTC)


I am very concerned with and opposed to giving blanket MEDRS endorsement to any book published by university press publishers as contemplated in the language above. For example, this book hardly looks like a thoughtful, mainstream discussion of environmental toxicology with it's self description as a "compelling expose" of the toxicity associated with household items such as brass door knobs and rayon scarfs. "No Safe Place" may be scientifically sound, but the introduction certainly presents it as an advocacy piece full of heros and villians rather than a dispassionate review of the data. Deceit and Denial "details the attempts by the chemical and lead industries to deceive Americans about the dangers that their deadly products present to workers, the public, and consumers." Please, please do not enshrine all these advocacy books as designated MEDRS-compliant sources Formerly 98 (talk) 18:52, 21 January 2015 (UTC)
Well, there will be the occasional exceptions, which is why the above language is "Books published by university presses or the National Academy of Sciences are generally reliable sources." I don't think that wording "enshrines" anymore than the current wording in the Books section, "Books published by university presses or the National Academy of Sciences, on the other hand, tend to be well-researched and useful for most purposes" - which is what I based it on. It sounds like you are basing your objections on initial impressions of the aforementioned books though. Blanc's book, for example, has been profiled in NCBI[2] and Scientific American[3] and would appear to be well-researched. I'm not convinced it's fair to assume it is an "advocacy" piece incompatible with MEDRS just because of a brief book description. --7157.118.25a (talk) 07:24, 22 January 2015 (UTC)
I don't think this is a useful change, and I'm at least somewhat concerned by what I perceive to be the motivations behind it. MastCell Talk 17:29, 22 January 2015 (UTC)
My "motivations"? I think MEDRS is overly obscure, is the most difficult policy to follow for new users, and thus has more potential for abuse than any other. It's in the best interest of Wikipedia to see it cleaned up. I have no other motivations than that. If you want me to speak plainly, the entire page is a wall of text that doesn't present key information concisely. It's overly wordy and in definite need of a clean-up.
Key information about how to follow the policy is not presented prominently, effectively, or in a timely fashion. The page might be suitable as a textbook section but in a wiki for usage by the general public I really think more summarizing is needed. As Albert Einstein once said, "everything must be made as simple as possible, but not one bit simpler." There is a lot of complex policy information on this page as well as detail on how to follow it, so it makes sense to simplify the page - assuming you want most people to be able to follow it of course. --7157.118.25a (talk) 04:44, 23 January 2015 (UTC)
My primary concern is just that information be simplified and condensed for better presentation, how that occurs is irrelevant to me, which is why I was not even going to provide a proposal until requested. --7157.118.25a (talk) 04:46, 23 January 2015 (UTC)
In general, I tend to be concerned by attempts to condense the contents of this guideline down to short, absolute-sounding lists of "do" and "do not" bullets. Wherever there is a statement in this guideline which can possibly be interpreted as as hard and fast, binary, yes-or-no rule, it tends to get (mis)used that way—generally to the detriment of the articles where it is applied. (See further up this talk page, for instance, for discussions regarding the problems that come from overreliance of the five-year rule of thumb.)
The problem of deciding whether or not published medical sources are both good enough for use in Wikipedia and represent the best available material is inherently a difficult task. It requires skill, judgement, and experience. Trying to come up with a checklist that mechanically produces thumbs-up/thumbs-down just doesn't work; the scientific literature isn't sufficiently uniform and well-behaved. What this guideline does well is provides a roadmap for how we think about medicine-related sources; the subheadings in the "Basic advice" section of the page represent a useful list of principles:
  • Respect secondary sources
  • Summarize scientific consensus
  • Assess evidence quality
  • Avoid over-emphasizing single studies, particularly in vitro or animal studies
  • Use up-to-date evidence
  • Use independent sources
(And the full text of that section is only a dozen or so mostly-short paragraphs—still quite digestible for someone who is interested in properly applying this guideline to Wikipedia's tens of thousands of medical articles.) The rest of the document in turn offers suggestions and rules of thumb for how those principles may be applied. Looking at the guideline as it stands now, I have to say that it just doesn't seem to be that bad, as long as you're not hoping to find a magic checklist. TenOfAllTrades(talk) 18:16, 23 January 2015 (UTC)
It's possible that we could improve the guideline by re-naming "Assess evidence quality" to something more relevant, like "Respect higher quality evidence". WhatamIdoing (talk) 19:26, 23 January 2015 (UTC)


The problem is that MEDRS when cited is rarely cited as a loose guideline but absolutely anyway, and is frequently used as a basis to make wholesale reverts to sourced material. From my experience the policy is often involved in edit warring, with new users simply told to follow WP:MEDRS without any instructions or details. See the above section where User:Sbharris similarly describes MEDRS as being used like a "cudgel." The way MEDRS gets used is not as a guideline or a general usage set of suggestions, but as absolutes; WP:MEDDATE for example gets cited to remove any sources older than 5 years.
Given that in practice it is being used to require much stricter sourcing, it's imperative that its standards be clear and concise so people can easily follow them. If MEDRS is going to be cited absolutely it should be presented clearly. Stricter standards on sourcing for technical subjects is perfectly justifiable - as long as you make clear what those standards are. My concern is that the instructions on this page are overly obscure resulting in few people being able to follow them, and ultimately leading to only a narrow subset of editors that are allowed to edit on technical subjects. Under such a scenario, I don't think that would meet the spirit of a public access wiki or a general usage policy.
Right now you have to do quite a bit of scrolling just to get past the lede and table of contents. Detail on how to actually meet these stricter standards is hidden in the very last page section in a vaguely named section called "Searching for Sources" which from the title is very undescriptive. It could refer to problems encountered while searching, ways to search, why you should search for sources, etc. That this section is actually key to meeting the MEDRS standards is hardly suggested by the section title.
Detail on what sources to use under the MEDRS standard is only mentioned in the second to last section, Choosing Sources, and then is not presented concisely. To get to this section, which by the way is also vaguely named, one must get through a huge, wordy section on Basic Advice before finding out what standards are actually required under MEDRS.
So again, if this page were an unenforceable guideline instead of authoritative policy on what medical sources to use, this wouldn't be an issue. But that is not the case. If this policy is to be broadly applied to the general Wikipedia community to govern all editing on medical and scientific topics, then key information about how to follow its exacting requirements should be clearly and visibly presented in a way most editors will understand. Otherwise the result will be edit wars because new editors don't understand how to follow an esoteric, highly obscure policy, and unknowingly violate a policy that is difficult to comprehend, resulting in good editors being needlessly removed from Wikipedia. --7157.118.25a (talk) 20:52, 23 January 2015 (UTC)
Ultimately, the issue here is NOT strict guidelines. The problem is overly broad wording that leaves open to interpretation what rules are concrete and what are not, allowing abuse. Editors can interpret a questionable, loosely explained facet of MEDRS as the basis for their edit warring; and revert material while forcing others to comb through an obscure policy to even find out what they are doing wrong. Again, stricter sourcing requirements are perfectly understandable for technical topics, but they ought to be presented clearly and specifically to avoid the kind of abuse associated with MEDRS which results in it being used like a "cudgel." --7157.118.25a (talk) 21:18, 23 January 2015 (UTC)

Can we please add guidelines about what qualifies as "medical"?

I propose that the rule of requiring medical sources should not apply to topics which medical journals currently treat as "out of scope". As examples, articles about cognitive psychology, philosophy of mind, religious counselling, or the "social health" of nations or businesses should not be held to the medical sources rule, even though medical journals may someday be willing to cover these topics as thoroughly as they do cancer and autism. Langchri (talk) 04:17, 27 January 2015 (UTC)

It's pretty simple. If an article has a statement that makes a medical claim, that statement (not the article as a whole) requires WP:MEDRS, regardless of whether it's an article on pancakes, the sugary goo that goes on top of them, miley cyrus, or pancreatic cancer. It's not the topic that determines a MEDRS-quality source requirement, it's the statement. What constitutes a medical claim should be blatantly obvious in nearly all cases. Medical articles often have history and culture sections which have statements that only require WP:RS simply because medical claims are infrequently made there. Seppi333 (Insert  | Maintained) 04:25, 27 January 2015 (UTC)
MEDRS specifies "biomedical", because some people think that everything is medical. We're trying to emphasize the "biological" part of it, rather than the social part of it. I've thought about writing a {{supplement}} (with examples) to better define WP:Biomedical information. The question comes up often enough that it's probably worth the effort. WhatamIdoing (talk) 06:54, 30 January 2015 (UTC)
Good luck. Is the statement that study A with herbal mouthwash B found greater reduction of gum disease than a placebo in a control group, a "medical claim"? Single PC RCTs are your #2 best grade of evidence given in this guide. Is it all unmentionable except where meta analyses or secondary mention can be found? Even the FDA requires no more. Are you saying this must be left out now, while folk beliefs about herb remedies can be retained as pharmaco-ethnology?SBHarris 01:16, 12 February 2015 (UTC)
Langchri, the majority of psychology topics involve peoples' health. Claiming that the human brain/mind works a certain way is a mental health topic, and I don't see why that should not be subject to WP:MEDRS unless it's some type of historical claim by a figure such as Aristotle. The WP:MEDDATE section of WP:MEDRS notes History sections as an exception. Any statement that has to do with health, unless it's a historical matter, should be subject to WP:MEDRS. There is a lot of information out there about bullying and its emotional effects, for example; bullying is no doubt a social topic, but making claims about its emotional effects should be subject to WP:MEDRS. I've stated this before, but the medical field is broad; everything listed in this section is going to be subject to WP:MEDRS at one point or another. Cognitive psychology involves how people think, and is therefore going to involve topics that are subject to WP:MEDRS. Flyer22 (talk) 12:08, 30 January 2015 (UTC)

Primary studies

I know, I know, secondary sources are always preferred when they exist. But a big journal just published results from a clinical trial and the findings were interesting. As I understand it, MEDRS says primary sources are allowed when reviews have yet to take place, correct? Then when the review does occur, we go with it only and remove the primary source, correct? LesVegas (talk) 18:42, 15 January 2015 (UTC)

Which primary source? Doc James (talk · contribs · email) 00:41, 16 January 2015 (UTC)
This is the sort of question that is best dealt with through discussion on an article talk page, or at WP:RSN (or WP:FTN, depending) if that doesn't work. The talk page of a policy or guideline isn't really a good place for rulings on the appropriate use of a specific source to make specific claims in a particular article. TenOfAllTrades(talk) 15:19, 16 January 2015 (UTC)
I'm sorry but that doesn't cut it when I have editors removing primary clinical studies and simply saying "per MEDRS". The MEDRS guideline has set itself up to be a superior epistemological standard over MEDMoS and WP:RS, and it had better carry the weight of that, or change itself to an "essay." SBHarris 01:29, 12 February 2015 (UTC)
User:Sbharris to what do you refer? I am happy to reply to User:LesVegas's question. It depends on the trial. For exception we discuss. Doc James (talk · contribs · email) 04:11, 12 February 2015 (UTC)
I reverted Sbharris' addition of a 'clarification needed' tag ([4]) a little while ago; I figure I should probably go into a bit more detail here about my reasoning. Unfortunately, the situations where we will use a new primary source, and the length of time for which we will keep using that primary source while waiting for secondary coverage, are going to be complex and varied. Some primary sources are going to be 'better' than others (even among the small subset of primary sources that are 'good enough' to be used at all in this fashion): larger studies, better methodologies, higher-impact journals. Some fields, medical specialties, and diseases are the focus of much more funding and activity than others; consequently some fields will see updated, appropriate secondary sources much more quickly than others. Unfortunately, it's not possible to put a specific 'expiry time' on primary studies, so it's not obvious how we woudl clarify that part of the guideline further.
As I've said elsewhere, the correct and successful application of WP:MEDRS relies (perhaps more than most Wikipedia policies or guidelines) on careful, knowledgeable interpretation. The suggestions that we already offer in the guideline (e.g. and especially the five-year rule of thumb for 'recent' secondary sources) tend already to be over-interpreted as hard-and-fast binary rules rather than as approximate guidance. (A review article from Feburary 2010 isn't magically good now and magically bad three weeks from now—but we regularly have editors who make that mistake.) Editors applying MEDRS are expected to use their judgement and are supported, advised, guided, and – if need be – corrected by a fairly dedicated group of editors who are familiar with both Wikipedia policies and the medical literature.
All that said, I think that there is room to revise MEDRS a bit in this area. While we should respect high-quality secondary sources, I'm not sure that we're entirely well-served by suggesting that primary sources should always be discarded and replaced by secondary sources (as they become available). In some instances, I think that we should try to retain the relevant primary sources alongside the summarizing secondary sources, and that this could be of benefit to our readers. (Consider, for example, a situation where a newly-published review article essentially summarizes and endorses the results of one or two clinical trials. In such a case, I think it would often be useful and constructive for us to add the secondary source as an additional footnote, but to still retain the original citations of the primary publications.) Just a thought. TenOfAllTrades(talk) 04:36, 12 February 2015 (UTC)
LesVegas has it backwards when they write "when reviews have yet to take place,..." The key words (used by TOAT) are: "alongside the summarizing secondary sources,..." We avoid standalone primary medical sources to avoid OR cherry picking. We can justify using such sources when it backs up their previous mention in reviews. It's a nice service to readers, and this doesn't apply only to medical content. Cherry picked primary sources can be problematic, so we must be careful.
The same principle applies in cases where we can justify using an otherwise terrible source like an obscure blog comment or Facebook comment when it has been been mentioned in a good secondary RS. For example, in the Robert Sears (physician) article there exist multiple secondary sources, including a CDC sources, which mention one of his patients as the index case in a measles outbreak, and we then supply his own confirmation of that fact in a reply to a reader's comment on his HuffPost column. We don't normally use reader comments as sources, but in this case it's not a SYNTH violation or improper because we have secondary RS mentioning the fact. Since there are readers who are confused about it, and we also have editors questioning it, it's a nice service to provide his own confirmation. -- Brangifer (talk) 15:44, 12 February 2015 (UTC)
I will note, to be clear, that LesVegas is dealing with the specific instructions surrounding the appropriately narrow use of primary sources in instances where secondary coverage in not yet available. Essentially, we know that review articles and other high-quality secondary sources will almost always trail the primary literature by some amount of time. If a primary result involves high-quality evidence (e.g. Phase III clinical trials versus mouse or in vitro studies) published in a high-quality venue in the recent past, then it may be appropriate to use the primary source directly in a Wikipedia article. We want to be particularly cautious when a new primary source makes a really remarkable or surprising claim, or where it contradicts previously-accepted results.
The window between primary publication and secondary review may be just a few weeks in a large, well-funded, 'sexy' field (sometimes there will even be commentary directly accompanying a primary publication for particularly high-impact work); the window period can stretch to some years in quieter subspecialties. To be honest, there are also instances where the review articles are actually lower-quality, less-trustworthy publications than the primary sources they cite. (This can happen when a grad student or postdoc needs a bit of CV padding, or when a kookier professor – often emeritus – cherry-picks some sources to promote a pet theory. Someone is friends with the editor of a backwater journal, and presto.)
As written, I find the guideline a bit problematic in that when secondary reviews come along we are instructed to drop the primary source altogether (even if it forms the principal basis for the secondary source's conclusions), per WP:MEDPRI:
  • "After enough time has passed for a review in the area to be published, the review should be cited in preference to the primary study."
There's actually a bit earlier in the same section which gives slightly different, somewhat contradictory, and frankly better, advice:
  • "A primary source may be presented adjunctively to a secondary source."
That is also often going to be the preferable course from a WP:V standpoint, as it provides readers and editors with both the original source material and the secondary endorsement side by side. (Special bonus, by the way, for the use of adjunctively in the guideline; we're clearly dealing with medical experts here.) I'm inclined to tweak the wording of the former passage to align better with the latter, if there aren't any objections. TenOfAllTrades(talk) 20:41, 12 February 2015 (UTC)
I like your thinking. -- Brangifer (talk) 02:33, 13 February 2015 (UTC)
I'm not convinced that those two statements are contradictory. It may be done, but you need to understand all the implications before you do it. This is using "should" in the RFC 2119 sense, not in the "I wrote should because someone told me that guidelines can't use the word must" sense.
I've got no love for the "give credit to the original" line of argument (which other people have proposed, probably including a reasonable proportion of people who publish primaries), and the secondary source may not constitute "an endorsement". Also, the secondaries that we like best are the ones that base their conclusions on far more than just one primary. WhatamIdoing (talk) 04:47, 13 February 2015 (UTC)
Your concern about "all the implications" is definitely correct, so we hope that any concerns will be mentioned here.
Unless I'm missing something, I think the context here (MEDRS) means that "secondaries" refers to reviews, unlike in a non-MEDRS situation. At least that's been my assumption. -- Brangifer (talk) 06:28, 13 February 2015 (UTC)

Pregnancy and lactation and drug interaction resources

Hello! User:Carly-pharm, User:C.G.Pharmacy and I are pharmacy students from the University of Waterloo. We are interested in updating antibiotic pages. What are Wikipedia’s trusted resources for pregnancy and lactation information? As well, what specific sources are recommended for citing drug interactions? Any feedback or suggestions are most welcome. Thanks! pharmerJAL (talk) 16:57, 6 February 2015 (UTC)

FDA and Australian government are good. Motherrisk is also suitable IMO as are review articles in pubmed indexed sources and the BNF. Doc James (talk · contribs · email) 19:00, 19 February 2015 (UTC)

It seems like I've seen a lot of references to the popular press section of this page. For such cases where it's directly relevant (as opposed to being mentioned when discussing MEDRS more broadly), does anyone have strong feelings about a shortcut? WP:MEDPOP? (It sounds like a terrible musical genre, but a Google for POPMED returns a lot of hits for population medicine). --Ryan (Wiki Ed) (talk) 18:53, 19 February 2015 (UTC)

(For context, this came to mind seeing SandyGeorgia's feedback to a student at Talk:Infomania. The student has the editing medical articles brochures and has been exposed to this, but clearly needed clarification regarding popular press sources). --Ryan (Wiki Ed) (talk) 18:55, 19 February 2015 (UTC)
Yes, please !  :) Best, SandyGeorgia (Talk) 18:56, 19 February 2015 (UTC)
 Done --Ryan (Wiki Ed) (talk) 19:06, 19 February 2015 (UTC)

RfC about appropriateness of medical sourcing

Watchers of this page might be interested in an RfC on Safety of electronic cigarettes regarding the appropriateness of statements released by medical organizations as sourcing for medical content. Arguments have been made that only published review articles can be used, and that tertiary sources are not allowed to be used on Wikipedia for medical content. Comments would be appreciated. Yobol (talk) 20:50, 19 February 2015 (UTC)

No one at all is arguing that tertiary sources are not allowed to be used on Wikipedia - that is simply a strawman. --Kim D. Petersen 21:50, 19 February 2015 (UTC)
Striked, and corrected. Yobol (talk) 22:05, 19 February 2015 (UTC)
The RFC is asking if sources like Press releases should be used on a medical page for medical claims. AlbinoFerret 22:50, 19 February 2015 (UTC)

Medical and scientific organisations

It was not obvious from the wording of this section that it would also include specialist academic centres of national or international reputation such as Memorial Sloan-Kettering Cancer Center, as well as government bodies and quangos. I have clarified this. Guy (Help!) 17:36, 30 March 2015 (UTC)

I don't think it should. The IARC, EMA, FDA, EPA, ACR, and ACP have dedicated groups that consider data, draw conclusions, and these are vetted before becoming official positions of the orgsnization. MSKCC has no such organized process that I am aware of, and statements "from MSKCC" are likely to be those of individual physicians, vetted only by the hospital press office. Formerly 98 talk|contribs|COI statement 17:50, 30 March 2015 (UTC)
Agree with Formerly 98; statements from individual medical centers' websites can be hit-or-miss as they may not be properly vetted as would a statement from a major medical organization be. This is to say, I view statements from individual centers' websites as borderline; often useful for uncontroversial statements, but better to use higher quality refs for anything remotely controversial. Yobol (talk) 18:11, 30 March 2015 (UTC)
Some academic sources can be reliable but a lot are unreliable. Editors can decide that on the talk page rather than suggesting they are all reliable. QuackGuru (talk) 18:21, 30 March 2015 (UTC)
Right now, editors are asserting that they are blanket not reliable because they aren't listed. In what way is a specialist academic centre of national or international standing, not reliable? Guy (Help!) 23:08, 30 March 2015 (UTC)
Maybe this might work. Proposed wording: "Reputable academic sources may be reliable, depending on the quality of the source." Thoughts. QuackGuru (talk) 00:04, 31 March 2015 (UTC)
It won't work, because those fighting quackery will always find a way to reject a source that does not serve their needs. The only thing that works is a set of pre-defined criteria to separate the reliable from the unreliable. -A1candidate 17:40, 31 March 2015 (UTC)
A list is impossible to manage. Proposed wording: "Reputable academic sources may or may not be reliable. This depends on the quality of the source." If it is carefully worded it might work. I'm not sure the best way to write this. QuackGuru (talk) 17:50, 31 March 2015 (UTC)
If you can't manage a simple list, see WP:CIR and read my previous comment again. -A1candidate 18:10, 31 March 2015 (UTC)
You're either drastically oversimplifying or don't understand all the nuances involved in determining the appropriateness of a particular use of a particular source in a particular context. Zad68 18:27, 31 March 2015 (UTC)
  • I have always argued that it would make much more sense to create a separate list of reliable medical sources based on several pre-defined criteria, rather than arbitrarily listing a source as reliable (or unreliable) based on whether it serves the needs of those fighting quackery. The MSKCC definitely is a reliable source due to its status as a designated NCI cancer center, but I could imagine the followers of Gorski and other SBM advocates desperately trying to bring it into disrepute as it speaks favorably about some CAM therapies per prevailing medical consensus. -A1candidate 17:33, 31 March 2015 (UTC)

BBC and Brussels Sprouts

I'd appreciate some comment, w.r. WP:MEDRS, on the following reverted edit [5] and its immediate reinstatement [6]. The content was removed with the summary "copyedit; rv WP:PRIMARY research; rv blog sources, WP:RS". This is not PRIMARY research or a blog, it's the BBC providing secondary reporting of it under their imprimatur, just as we look for. Also the Science Daily website, which I'm unfamiliar with, but again it's rather more than "a blog".

Looking at Special:Contributions/Zefr there is a rather obvious pattern here. Bulk removals of sourced content, claiming the sources are inadequate, with plenty of edit-warring to strong arm the changes into place. Looking at Ginger we see [7] where work being published out via PubMed is getting removed as "rv WP:PRIMARY, not WP:MEDRS".

I'm no medic. I have no idea if Zefr is either (their userpage is silent). However I see their pursuit of source pruning as going way beyond any sensible imposition of sourcing policies, particularly when it comes to stripping sources for being too primary alone. That is not the purpose of WP:PRIMARY, when that primary material is being published via the route of credible academic bodies and peer-reviewed journals. Andy Dingley (talk) 22:27, 22 April 2015 (UTC)

Hi Andy Dingley,
BBC News (typical articles, including this one) is an independent source but not a secondary source. A secondary source is an intellectual work that does more than just repeat what someone else says by transforming it through critical analysis. See WP:PRIMARYNEWS for more details (spoiler: it's complicated, because some news sources really are secondaries). WhatamIdoing (talk) 19:40, 27 April 2015 (UTC)

Veterinary medicine

Does, or should, MEDRS cover Veterinary medicine? I have seen twice in the past week editors write that MEDRS may be applicable. I can provide diffs if necessary, but these comments were asides to other discussions, and the main issue here is whether animal health is within or outside of the scope of MEDRS. If this were made explicit, we could avoid future arguments over animal health sources.Dialectric (talk) 23:20, 8 April 2015 (UTC) @Dialectric: can yo provide the diffs in question? Montanabw(talk) 19:59, 17 April 2015 (UTC)

There have been some slight discussions on this in the past, but the main one is here: [8]. Personally, I would say yes (or mostly) coming from a veterinary background. The only difference between the field of veterinary medicine and human medicine for us as editors here is that animal models aren't as much of an issue because the animal can be tested directly without the ethical issues of testing humans for certain things (e.g. lethal treatment groups, etc.). Beyond that, what MEDRS describes is pretty much parallel for the medical sciences or really science in general. Reasons why we as editors avoid primary studies for instance are pretty ubiquitous across disciplines. Kingofaces43 (talk) 23:34, 8 April 2015 (UTC)
Thanks for the link - that discussion was a good start. Your bolded question 'Does MEDRS apply to non-human branches of medicine? If not, why?' is similar to what I'm asking here, and there are a few interesting questions raised about the bounds of MEDRS.Dialectric (talk) 04:55, 9 April 2015 (UTC)
I would be against WP:MEDRS as it currently exists being applied to veterinary medicine (vet med). At the moment, WP:MEDRS contains sections called -
1) Avoid over-emphasizing single studies, particularly in vitro or animal studies
2) Use up-to-date evidence (<5 yrs old)
The first will be difficult to apply to vet med for several reasons. Vet Med often depends on single case studies (e.g. surgical techniques, diseases of zoo animals). These are published, but are very unlikely to be subjected to further research; their efficacy is subsequently reported and tested by word of mouth, electronic noticeboards, etc. but not in the scientific media. This first section is also problematic because it advocates not using studies on animals. Of course if a vaccine is being developed for cats, it should be tested on cats. This is a relatively minor point - I think a simple wording of "target species" could cover application to both humans and non-human animals.
Section 2 (up-to-date evidence) is likely to be problematic due to the nature of research in vet med. Once a practice, procedure, whatever, is established, it is extremely difficult to get further funding for this. Therefore, over time, the sources will slip past this 5 year cut-off (where does this come from anyway?). I have never seen guidelines applying a "recency" component to vet med writing.
I am not saying WP:MEDRS can not be applied to vet med, but I feel that as it stands, this would cause problems. For example, the section on suitable sources discusses the strength of experimental design/sources. It states that "meta-analyses of randomized controlled trials (RCTs) are the best design/source". It then describes the other exptl designs almost as if they should be never be cited. They are probably out there somewhere, but I can't ever remember seeing meta-analyses of randomized controlled trials in vet med. Unfortunately, an over zealous editor could quite easily use this to prevent an article from progressing. I also agree as suggested above, it would need to be clearly stated what is the scope of vet med or perhaps even veterinary science. Just thoughts - I hope these help.__DrChrissy (talk) 15:04, 11 April 2015 (UTC)
Not too problematic tweaks indeed. The caveat with 1. is using findings from one species to infer something about another. That's the real meaning of the animal studies point, so it might actually be good to be specific about that anyways. For 2. that's really meant to say use the most recent studies available. 5 years is a decent cut-off in most sciences if reviews come out very often, so it basically means newer than 5 years, if none are available then use the most recent. I don't think availability of reviews is a problem though. I just checked Web of Science for reviews using "veterinary" alone in the last 5 years and got 608 reviews. I have never run into issues finding a review for veterinary topics I'm interested in either. Overall though, the spirit and scope of MEDRS would seem to be appropriate for veterinary med with some slight clarifications, so it would seem simple with slight clarifications. Kingofaces43 (talk) 01:45, 12 April 2015 (UTC)
  • Comment: Per the comment above, WP:SCIRS already does apply. I take no position for or against looking at bringing in MEDRS, except that given how some people treat MEDRS with extreme literalness, I'd suggest having a specific subsection for veterinary medicine that addresses the very legitimate concerns that DrChrissy raises. Also, because alternative medicine is more openly accepted in the veterinary world, we may want to note that some things that would not fly in human medicine may be acceptable in a veterinary article (so long as there is adequate sourcing) - for example, many veterinarians also do animal chiropratic medicine or acupuncture. Montanabw(talk) 22:39, 12 April 2015 (UTC)

Question The article Specialty (medicine) lists psychiatry as a specialist topic. The Psychiatry article defines this as "the medical specialty devoted to the study, diagnosis, treatment, and prevention of mental disorders." There are analgous processes in the veterinary world. Are people arguing to use WP:MEDRS on articles such as Dog training, Clicker training, Obedience training, Separation anxiety in dogs, etc, etc? This is not a rhetorical question.__DrChrissy (talk) 10:59, 13 April 2015 (UTC)

  • I think that animal training is closer to education as an analogous comparison to humans. By and large, I think that falls under a "don't scratch this if it doesn't itch" category. (And if some idiot is arguing for MEDRS, point them to SCRIS instead. Montanabw(talk) 05:13, 14 April 2015 (UTC)
    • I agree with you totally. I was thinking more about animal behavioural therapy rather than training per se. My concern here is that imposing WP:MEDRS on Vet Med, especially without carefully defining the topics, will wipe many articles from the project and cause an enormous amount of disruption. For example, the article describing a piece of veterinary medicine equipment, the Elizabethan collar, would not survive even the lightest of touches with WP:MEDRS, but currently reads sufficiently well for the average reader. If people want to get a flavour of articles, perhaps look at the categories pages. Category:Veterinary medicine articles by quality states there are 1,072 articles in this category (I have no idea how up-to-date this is). Category:Veterinary medicine has 22 sub-categories (YIKES! I have just seen Winged cat! That needs to go!)__DrChrissy (talk) 08:40, 14 April 2015 (UTC)
Forgot to reply to this, but for an example like Elizabethan collar I don't really see anywhere that MEDRS would apply. Basic equipment normally doesn't fall under a need peer-reviewed sources. It's when you start getting into details about a specific treatment that you'd really start digging into vet med journal articles. For instance, over at trocar, I wouldn't really think MEDRS sources are needed to describe what it is and what it is used for. How well it actually works for dealing with bloating cattle and survival after using it would be the specific piece of content I'd look for journal articles on. I believe this is how most medical equipment articles are currently handled as well. Kingofaces43 (talk) 00:37, 16 April 2015 (UTC)
A lot of thoughts here, so bear with me. Training type articles would be describing what is done in training and doesn't really fall into the realm of any particular science (I'm not aware of journals covering something like this). That would be equivalent to say an article on human parenting methods where we don't really apply MEDRS per se. Saying whether a particular method is effective really does require scientific studies though. Once you start getting into actual health content though (i.e., diseases, disorders, nutrition, etc.), those are areas where there are plenty of scientific publications where we should be pulling from our best available sources (typically review articles). This is the area I see clearly within the scope of MEDRS because we are pulling from exactly the same kinds of studies in either discipline with the tweaked caveat about extrapolating findings from model organisms to the species of interest (human or otherwise).
That all does start going towards the question I asked here though about SCIRS. After wrestling with that question for awhile I would take the MEDRS approach for obvious health content (i.e. reviews and reputable organizations) because we as editors still should not be interpreting the findings of primary studies. As you move away towards more tangential topics that are not so much about health, but maybe general social behavior of wolves, chickens, etc. you still have scientific research there, so I'd look to SCIRS still looking to secondary sources, but acknowledging we might need to look to the introductions of primary sources in place of formal literature reviews with the cautions that come with that. Behavior is one topic that can straddle the line between MEDRS and SCIRS, so it really depends on what a specific piece of content is about.
So, I would put all of the training articles you listed as generally under SCIRS since there is a mix of topics in play. If one was to comment on how effective training methods are, that’s where I would be looking towards scientific topics specifically, but otherwise general sources are fine for general descriptions. The separation anxiety article would clearly be a health topic though, so I would be looking to journal articles for information there as opposed to people who just write about their ideas on it. The take home message is to pull from our best available sources regardless of MEDRS, SCIRS, etc. labels. Kingofaces43 (talk) 20:04, 14 April 2015 (UTC)
  • Question: Why is this even being raised? Is there an actual problem somewhere? Otherwise, we may be discussing solutions in search of a problem! If nothing itches, I advise not scratching. JMO. Montanabw(talk) 06:55, 16 April 2015 (UTC)
If you worked on bird, frog, or bee articles, for examples, and the ways that various chemicals may be affecting them, you would understand. IMO. Gandydancer (talk) 07:41, 16 April 2015 (UTC)
I do work on articles like that and I believe User:Montanabw also does (please correct me if I am wrong). Let's talk about the content, rather than the editors, please.__DrChrissy (talk) 08:50, 16 April 2015 (UTC)
To be clear, the editor states: "... the main issue here is whether animal health is within or outside of the scope of MEDRS. If this were made explicit, we could avoid future arguments over animal health sources". Suggestions have been made that MED:RS should apply to animals as well and I am answering to that question. I do work in those areas (bees, etc.) and I have not noted Montana's involvement though certainly I could be wrong. Gandydancer (talk) 13:08, 16 April 2015 (UTC)
My question is if there is an article within the scope of WP:Veterinary medicine where this issue is actually being raised; it should be noted here so that we can be discussing ACTUAL problems and not theoretical ones. @Gandydancer: I work on some articles within the scope of WP:Veterinary medicine (for example, I created Colitis-X, and Parascaris equorum). Knock off the condescension and personal attacks. Montanabw(talk) 16:53, 16 April 2015 (UTC)
Good grief, I had no intent of being condescending or attacking you! I always thought that I liked you and the work you do here and I've always considered you to be one of our best editors. I'm aware of the work that you do related to horses. I just wanted to point out that if we begin to insist of MED:RS for animals, animals include the health of birds, bugs, frogs, etc., which have of late seen a lot of connections to the ever increasing use of chemicals and their demise. I've already been told to leave chemical articles by chemists because they are supposed to be about the chemical and not about the way that a chemical is used when it is used as a pesticide and that this or that study is not acceptable because it is not MED:RS approved. I have no intent to argue here and am sorry that I said anything. Gandydancer (talk) 17:14, 16 April 2015 (UTC)
Is there an actual problem in these topics though? We generally rely on reviews to interpret science sources in general (i.e., we're not qualified to assess the actual findings of primary sources). When it comes to chemicals and pesticides, reviews tend to come out pretty often summarizing the literature, so I'm not seeing any particular problems coming around by taking a more MEDRS approach there. The issues with primary journal articles and why we avoid them are generally the same across disciplines, so what distinction are you seeing here Gandydancer? I personally don't see any issues on the pesticide front. Kingofaces43 (talk) 19:00, 16 April 2015 (UTC)
The problem I see here is that adherance to secondary sources has not been as stringent in science articles outside of what is currently covered by WP:MEDRS. In my own editing of animal related articles, I very, very frequently use primary sources. TuT TuT I hear you say ;-). But, I follow with strict 100% adherance that content is verifiable. There would also be a problem with newspaper articles. WP:MEDRS views these as primary sources, but in other subjects they are viewed as secondary sources. In some subjects, newspapers might be the only source of information (e.g. attacks by animals, spread of a disease). I appreciate that in the human medical world it is critically important to have the most robust reporting of the appearance and spread of a new flu virus, but do we need the same adherance to report that "In 2014, an individual of species X was seen in location Y for the first time - as reported in The Guardian"?__DrChrissy (talk) 11:32, 17 April 2015 (UTC)
I'll be away from any computer for the weekend, so I'll have to respond a bit more later. The problem with using primary sources is using their results or conclusions. I'm fine with using the introductions carefully when literature reviews aren't available, but we are no in position to use the findings of primary studies. That's the main thrust of MEDRS I'm looking at here. Newspapers articles generally wouldn't be reliable because they don't do a good job reporting on science with respect to research, but mundane things like a dog bite or reports of a disease occurring aren't exactly cutting edge science. There's a big difference between what's within the scope of journals in terms of research, and more basic biology that can be described by any number of sources. For instance, your "MEDRS" example here would be taking MEDRS a bit too far. Most of your secondary source required pieces of content do not need a review because they are just basic biology that are reported on in other acceptable sources such as extension publications, journal article introductions, etc. That's more SCIRS-y kind of content there. Content that requires more in-depth knowledge, such as your "The honey bee's venom . . ." would need some commentary form an outside source more in line with MEDRS though.
Basically, if it's basic biology, we'll have plenty of sources to pull from that are still reliable even if they aren't journal publications. When it comes to describing research that's more in-depth though, primary sources don't really work for us here, and that's where the general concepts behind MEDRS kick in. I'm not sure if you were picking that up reading over that example you made, so is that distinction making a bit more sense now? Kingofaces43 (talk) 17:01, 17 April 2015 (UTC)

Seems to me that WP:SCIRS covers most of what is needed. Montanabw(talk) 19:59, 17 April 2015 (UTC)

In that particular case yes. It doesn't really get into the realm of vet med though. Kingofaces43 (talk) 21:07, 17 April 2015 (UTC)

Meshing of MEDRS with animal studies

Here's a scenario I've been thinking about. Let's say someone is soapboxing on a particular article cherrypicking primary studies on a crusade to make chemical X look really good/bad, exposing the "truth", etc. In terms of content in the context of human health, primary animal studies would not be appropriate, and that's one of our guards against more advocacy in medical topics. Instead, the editor says the content is not about humans (not delving into whether that's a good faith intent or a slightly wikilawyerish move), and the primary source should be fine. Let's say it's a rat study. If someone is really fine with primary studies when humans aren't being mentioned, we technically could fill medical articles with animal studies at that point. It's a little bit of a potential loophole (and a slippery slope), so how would others deal with this example?

Personally, I'd be looking for context of whether the study was really veterinary in context (e.g. chemical X causes problems for livestock production/pets), or if it was just a general animal study just used as a model organism. The former would be veterinary where I'd still personally be looking for a secondary source to summarize the particular malady or treatment, whereas the latter would fit more squarely in what we don't include under MEDRS with regard to model organism studies. Essentially, what field does the weight originate from? Does that seem like a reasonable approach? Kingofaces43 (talk) 17:54, 5 May 2015 (UTC)

I think that's a great place to start. WhatamIdoing (talk) 19:21, 11 May 2015 (UTC)

Seneff redux

Don't know if folks here remember the Seneff article on glyphosate that caused some ruckus a couple of years ago, which we discussed here and found unreliable for making health claims (see here) but there is another review published in 2013, equally FRINGEy. See PMID 24678255. Apparently this has lit up the blogosphere. See the comments at the bottom of the abstract, which provide links to "Oh, no! GMOs are going to make everyone autistic!" and "Glyphosate – The New Bogeyman" at Science Based Medicine. In my view PMID 24678255 is not reliable for making any kind of health claim. Comments? Jytdog (talk) 15:06, 4 May 2015 (UTC)

the journal is not MEDLINE indexed. Publisher is Bratislava : Slovak Toxicology Society SETOX : Institute of Experimental Pharmacology & Toxicology, SASc.. Impact factor of 0, [per research gate. And see this about the journal as well. Jytdog (talk) 15:30, 4 May 2015 (UTC)
The thing that has been conclusively proven that GMOs cause is cancer, not autism, though chemicals used on GMO crops do cause brain damage. The extremely cancerous compounds in GMOs were proven by a French scientist who's first name is Nick, and of course in France, since it is not America, this was front-page news. The study was then incorrectly redacted from the journal due to a frivolous complaint by Monsanto, stating that "the rat species he used is prone to cancer", regardless of the fact that Monsanto's own "experiments" which they claim to prove that GMOs are safe, used the same species of rat, as well as regardless of the fact that the GMO rats did get massive amounts of cancer while the organic-fed and non-organic, non-GMO -fed rats did not. There was great protesting by mass amounts of scientists to the magazine, and the journal eventually reinstated the study a few years later, admitting their wrongdoing. This was reported by Abby Martin at Russia Today (a source that EVERYONE agrees is reliable) on the week that it was reinstated. Obviously Agent Orange (glyphosate) does cause brain damage, but until the opponents are aware of the fact that autism and chemical brain damage will be very similar in appearance in many cases, it is not autism that is the main "speaking point" regarding GMOs. As for Dr. Seneff, I don't know about her articles, but there is a 2-hour lecture which explains in detail her information, which is easier to follow the scientific information on then an article is likely to be, because it is very complicated.                     ~Rayvn  18:40, 12 May 2015 (UTC) — Preceding unsigned comment added by RayvnEQ (talkcontribs)
almost everything you wrote is wrong. oy. Jytdog (talk) 18:48, 12 May 2015 (UTC)