Health as a topic is a bit different than what most SE sites deal with. It is a lot more likely to attract users without a medical background than a programming site is to attract non-programmers. Many people have questions about health, fewer are qualified to answer those.

The consequences of bad health advice can also be rather severe, and while no sane person should just ask the internet for serious issues regarding health instead of their doctor, people will inevitably do so.

Some sites like Skeptics have strict requirements for answers regarding references, any answer that doesn't substantiate its claims with appropriate references is removed. The reason on Skeptics is that the site allows pretty much any topic, so relying on voting alone doesn't work as most users aren't an expert in the topic of any question.

Should we have a similar requirement for answers to back up their claims with references from reputable sources like medical journals or trusted medical institutions?

  • 2
    Comments are also frequently used to dispense terrible advice, like this one. What do we do about this? Commented Apr 5, 2015 at 18:07
  • @anongoodnurse, what comment are you referring to? Has it been deleted?
    – Garrett
    Commented Apr 5, 2015 at 22:28
  • @anongoodnurse was that the binders comment?
    – Tim
    Commented Apr 6, 2015 at 11:12
  • 1
    @Tim - I don't want to make a fuss; the commenter was exceedingly gracious and removed it. The point still stands. Bad advice can be dispensed in comments as well as answers. It's a problem we should be aware of, and maybe flag more often (?) than on other sites for. The problem is: who knows it's bad advice? Commented Apr 6, 2015 at 11:23
  • 1
    @anongoodnurse Yes, I am concerned about the same thing. I think that discussions like that should be kept to chat and answers should be answers.
    – Tim
    Commented Apr 6, 2015 at 11:41
  • @anongoodnurse Great question! You should ask it as a new question. Commented Apr 9, 2015 at 12:27

8 Answers 8


I vote YES. Absolutely.

That's the whole point IMHO; otherwise this site will fall on an equal not-so-sweet spot between usefulness and uselessness as yahoo answers.

The only way we're going to move away from the abundance of "voodoo-medicine" is by demanding factual evidence. If it's not documented, it is not medicine, it is quackery, even if it sounds right because "common sense".

I didn't go through the Area 51 process of voting, following, committing, and signing up to this site to be able to access "fairly proven common knowledge".
I came for the scientific health info exchange; I came for scientific data & knowledge translated to health advise when possible, if you may.

This site offers an opportunity to reduce the amount of disinformation and misinformation that prevail out there in every culture because of voodoo medicine and quackery. Patients are sick of it, doctors are sick of it, only charlatans profit. More and BETTER information will lead to healthier people and more doctors with enough time on their hands to pursue research efforts and push forward on medical discovery and progress.

Remember 2 things:

  • Stack exchange sites strive to be for experts by experts, so the non-experts can read and learn actual facts and truly useful information.
  • Stack exchange pages must have enough info to stand on their own and not rely so much on external sources because the web changes and links may break; so you should link references and also cite the relevant textual content if possible (unless copyrights don't allow it).

And let's not forget this topic is not arcade/gaming, movies & sci-fi where "mistakes" are almost (arguably) meaningless in the real world. This topic/site importance is comparable to (or more important than) programming, programmers and such, where not having access to the right information can cost jobs, money or even lives.

So, everything must be taken with a grain of salt? Yes, always, but that is no excuse for mediocre questions/answers.

I would very much like it if this site grew as much as stackoverflow or at least as big as serverfault, superuser & askubuntu. I'm not a doctor so I've been inviting all the reputable and respectable doctors I know. For starters my brother who is young but amazingly talented on both his intellect and his way of explaining complicated processes in a simple to understand manner.

I've seen Computer Sciences grow so very damn fast in a great deal thanks to the open and ever-generous community. I owe you guys a lot; you help me and I help you. We share and share knowledge that's easy to debug and test.

Our health deserves the same love.

Our health depends on the correct function of a system made of subsystems that were not designed by us. So, regardless if you believe it was intelligent design or evolution, the fact is that all of our solid knowledge of this system has come from reverse engineering human biology or another kind of biology as a mean to understand our own.

I will be down-voting any answer that does not provide ANY kind of reference (and it should not be taken personal) that's lazy, plain and simple. It doesn't even have to be a totally irrefutable source; it can range from believable to reputable to trust-worthy to actual evidence. The point being that experts should exchange ideas & information with a problem solving mindset; vetted by experts, half baked answers with bad ideas and dubious references would sink to the bottom and the best ideas would rise to the top and we should know where those ideas came from.

So yeah, you should most definitely add reference links and cite sources, otherwise we have to take only your word for it. Taking your word is OK if we're in a hospital or a clinic because we can expect the person talking on a white coat to be a qualified professional, but on a web forum not so much, and if not providing references becomes accepted as part of this SE site's culture, then the site becomes pointless and useless to me and for you. We could very well be reading a consumer magazine (reader's digest) or a fashion magazine article and call it a day; but I don't want that, I want the evidence, I want the science.

  • 7
    +1 - I think this is essential to our usefulness. Medicine is also a field where some people enjoy demonstrating (a non-existent) expertise. We should be able to read the reference if we desire. Commented Apr 4, 2015 at 7:27
  • 4
    I agree. Including a few references may not be the be all and end all of medical discussion, references are only one critical requirement for us to take any opinion/argument seriously to begin with. We need to know where anyone's ideas come from to test how solid those ideas are. Commented Apr 5, 2015 at 1:31
  • health.stackexchange.com/questions/ask/advice? does not say "Stack exchange sites strive to be for experts by experts" , does it?
    – qdinar
    Commented Mar 1, 2018 at 16:31

I think the approach should be that we "strongly encourage" references to reliable web sources in answers (as, for example, the Skeptics Q&A does), so you're not just taking someone's word for it when it comes to medical advice/info. Perhaps if a thorough enough answer is given, a reference wouldn't be required, but usually it would.

  • 2
    I think Skeptics.SE is the model here. If we can maintain the rigour they do on their site both for questions and answers I think we'll be on a sure footing.
    – Tom Medley
    Commented Apr 7, 2015 at 11:01

While there is no specific requirement, on the Fitness SE site we deal with it by downvoting "bro-science" answers or challenging the person to back up their claims with data.

I think that if you are relating something that is fairly proven common knowledge, you don't necessarily need to cite sources, but if you are posting something that goes against "common" knowledge or in an area where there are conflicting answers that backing it up would be required. Otherwise you run into a "he said/she said" type of situation, and people upvoting whichever position they happen to agree with.

  • 2
    "fairly proven common knowledge" sounds like an unacceptably subjective term. Additionally, as HealthSE grows it will be harder and harder for a handful of mods to determine accurately every single post's content. Unless you have consistently encouraged users to flag non-referenced posts, you ll end up with too much to handle on your own. I strongly suggest not rejecting flags that flag a post as having "no references". Finally, it would be very disheartening to experts if they see bad content not downvoted/locked/marked.
    – user
    Commented Jan 6, 2016 at 18:44

In short: the current moderation policy is that answers with no references are acceptable.

enter image description here

  • Please note that there are many answers that don't have references. They are not all being removed. References are desired but not required (see top-voted answer.) Flags should not be used for wrong answers or those that have technical inaccuracies; they should also not be used for trying to prove a point you may want to make. Finally, he happens to have answered that correctly. An advantage I have as a physician: I can easily pick out a correct answer from an incorrect one. I would rather have a correct answer without a reference than in incorrect one with a bad or misinterpreted reference. Commented May 14, 2015 at 2:33
  • @anongoodnurse I wasn't sure that the top-voted answer was the enforced one, hence the flag. How else would I know what's the enforced policy? Commented May 14, 2015 at 3:04
  • There is no "enforced policy" here that I know of, except that the question should be about health. Otherwise it is like many other SE sites; questions and answers are evaluated - by the community, hopefully - on their individual merits. Commented May 14, 2015 at 3:08
  • @anongoodnurse I believe that one example of enforced policy on this SE is that in answer, not matter its individual merits, shouldn't exclusivity composed of references and quotations. If so, it should be deleted. Hence my belief that there could be an enforced policy regarding answers with no references. Commented May 14, 2015 at 3:11
  • I would agree that such answers are not good. However, I'm not sure if that means there is no such answer on this site (I haven't read every answer here.) I do pay attention to those I come across, like everyone else. Has the concern in your answer not been addressed? Commented May 14, 2015 at 3:15
  • @anongoodnurse Shog9 deleted all answers he saw that had references only: meta.health.stackexchange.com/q/193/43 . And, I have just learnt it, I see that Shog9 asked people to flag all answers with no reference, or answers solely containing references. So I'm still quite confused. Commented May 14, 2015 at 3:28
  • I have read @Shog's post and see no request to flag answers without references. Can you please provide a quote? Commented May 14, 2015 at 3:34
  • @anongoodnurse "Flag answers that speculate but don't back it up. Any good answer here should be able to draw on a wealth of available sources; failure to do so casts doubt on its validity and creates more work for anyone reading." Commented May 14, 2015 at 3:39
  • Yes, I saw that. I apparently interpret "speculation" differently than you do. For example, when I give patients answers, I do not provide references, yet it's not speculation. As I said before, that answer is correct. This answer, I believe, is speculative. Again, I'd prefer better, but I'd rather a correct answer without references than an incorrect one with. Commented May 14, 2015 at 3:55


This site is not FDA. It's not its job to protect people against themselves. It's not even possible to do so.

  1. Some answers will come from first-hand experience. E.g. the observation that glaucoma drug (Bimatoprost/Latisse/Lumigan) promotes eyelash growth. Requiring references to back up such claims would put us at same disadvantage as pharmaceutical corporations - having to wait years until trials are completed and reference material created.

  2. Some people will come here exactly for "fairly proven common knowledge", just like on lifehacks .

  3. My most important argument is that there is plenty of bogus reference. Infamous "vaccines cause autism" claim did have reference of nothing less but The Lancet itself. This alone, in my opinion, should shatter any faith that references alone can make this site reliable.

    By requiring links to references, we're encouraging users to trust potentially malicious references provided by potentially malicious author. I believe we should rater encourage askers to research answers on their own instead of trusting them blindingly.

  4. Skeptics is the worst role model possible. Skeptics is not a site for "pretty much any topic" that couldn't fit any other Stack. The very reason of Skeptics is to bring up some "fairly proven common knowledge" in order to confirm or disprove it. This makes references the whole point of skeptics, which is absolutely not the case here. We should stop comparing Health to Skeptics, because their goals couldn't be more different.

There is certainly risk of bad answers. I propose deleting answers proven wrong rather than deleting answers unproven right. Perhaps not even deleting, but leaving disproved answer with correcting notice can be as helpful as good answer.

I strongly believe that all humans instinctively take health topics far more serious than any others. But this is the instinct, and like many others, it no longer serves us right. Outdoors, Travel, Home Improvement, Motor Vehicles, Engineering - all these also carry (potentially life-threatening) severe consequences (and are not experts-to-experts). We should take this irrational fear into account and force ourselves to stop treating Health with undue reverence. After all, every drug OTC or not, carries it's own pamphlet of warnings. Users already have access to those, what we need is to remind to read them, not make an impression that "you're safe, we've read them for you".

There is no need of holding Health to higher standards than other Stack QAs. There is no need of trying to act as professionally as doctors.

//edit: This site is already too focused on the references while ignoring actual knowledge. See comments under my answer: Is Low Blue Light helpful for preserving eyesight? One guy mentions "They cite a lot of studies" on some website and moderator immediately goes "That would make a great answer". There is no attention paid to actually answering the question neither to the quality of it. Only immediate jump to rushed conlusion: "many references = great answer". We need to stop this mindset!

  • update on 30 Sep: the current policy is that answer without reference will have a post-notice
    – Ooker
    Commented Sep 29, 2015 at 20:52

I think the answer should be No, because it otherwise reduces the value of this site to nearly zero - it becomes a mere link collection.

If only answers with valid references are allowed, there is no gain for anyone in using health.stackexchange; people can just google (or bing) the question and find the references themselves. If there are no references to be found, then no answer would be allowed here either, so the question would just sit unanswered forever.

I seriously wonder why so many of you think 'Yes' would be a good answer - what sense does the whole site then make?

I think a lot can be gained for the asker by qualified answers without a reference, even though that leaves of course some risk that the answer is incorrect (but a reference leaves that risk too)

  • 2
    People just using Google or Bing are actually unlikely to really find the answer unless they know exactly what they are looking for (for example, what terms to search for), and is much more likely to encounter false inform. A proper search of scientific literature and being able to evaluate whether what comes up is relevant and scientifically sound is not that trivial. I think you are confusing asking for scientific references with asking for any kind of reference.
    – YviDe
    Commented Dec 29, 2015 at 11:04
  • "I seriously wonder why so many of you think 'Yes' would be a good answer - what sense does the whole site then make?" The whole site only makes sense if it is giving good advice. The only way to know if the advice is good is if it's backed up by reliable sources. If someone asks about best treatment for stomach ulcers, and a user answers one large clove of garlic chewed thoroughly and washed down with 2 Tbs. hydrogen peroxide five times a day until asymptomatic, how in heaven's name does this site make any sense? Commented Dec 29, 2015 at 19:55
  • @anongoodnurse, but you didn't address my point - the site wouldn't give any advice, it would just be a list of references which give good advice. Nothing of value gets added, because it's not allowed - all facts must come from the reference.
    – Aganju
    Commented Dec 29, 2015 at 20:37
  • 1
    Why do you say the site doesn't give advice? There are many answers already which summarize relevant literature. This answer for example, summarizes the literature and gives advice. If you mean something else, I don't understand what you're trying to say. If you're referring to personal experience as a source of advice, refer to my previous comment. That's considered anecdotal, thus unreliable. Commented Dec 29, 2015 at 20:53

A big problem here is that from only the peer reviewed scientific literature you'll have a difficult time extracting what a healthy lifestyle is. Most of what we know is healthy was known for thousands of years, and only some of this has been rigorously proven.

Fundamentally the problem is that there is big gap between being physically fit and ill. Health and fitness is not merely "not being ill", but the medical literature is mostly focused on illness. It is just like your new car that can drive 200 km/h and accelerate from 0 to 100 km/h in 3 seconds that is not merely "not a wreck". The knowledge on how to repair damaged cars won't help a race car driver all that much to fine tune his car to win the next race.

So, I take it the downvoters would disagree with the mother in the following conversation?

Mother to son: "Eat the vegetables, broccoli is healthy."

Son to mother: "I don't like it, it doesn't taste good."

Mother to son: "Just eat it, it's good for your health."

Son to mother: "Prove it."

Mother to son: "What?"

Son to mother: "That's what the teacher explained in class today. He said that when it comes to statement of facts, you can't just say something, you have to prove it. That's how science works."

Mother to son: "Look, just eat your broccoli and shut up, or else you won't get dessert."

  • That's rather random. It's also very faulty logic. Commented Jul 11, 2015 at 4:04
  • @anongoodnurse In case of lack of knowledge (when little or no conclusive research results exist) we should give the benefit of the doubt to our body design. E.g. knowing that we've been around for more than 200,000 years and most of that time we would have had far more exercise than today, it's a safe bet that many processes have to some degree become dependent on regular exercise. So, living a couch potato life will likely negatively influence almost all processes to some degree. Commented Jul 11, 2015 at 4:17

There are many questions that don't have any evidence and any past literature. So: No.

You must log in to answer this question.