12

Inspired by the comment left for my answer to this question: Can getting the flu shot make you sicker next year?

Providing sources is within the spirit of the site, and something I've tried to do when possible, there are going to be questions based on pure conjecture for which finding a source is non-trivial, if not impossible. I'd assert that this question is one - I've been working on influenza for the better part of a decade, and never even heard this assertion, let alone seen a study that might address it. After some brief Googling for the original belief, and then some Google Scholaring/PubMedding for a source to refute it, I came up with nothing and hit the "Stack Exchange doesn't pay me" limits of my patience.

How are we going to handle these types of questions. For example, if I post "Can taking beta-blockers turn your skin purple?", can we really expect someone to come forward with peer-reviewed literature to refute that? It's both an unstudied problem and attempting to prove a negative.

  • my complaint wasn't "your answer is only good, why isn't it perfect?" but you said "flu doesn't work that way", well then how does it work? What is the difference between flu and Dengue? Why should I believe a random statement on the internet that my friend's belief isn't true when you've done practically nothing to explain it? – Zaralynda Apr 17 '15 at 2:10
  • 1
    @Zaralynda I said flu doesn't work that way. Gave an example of a virus that does. Noted that none of the immune dynamics of influenza match those of a disease where that would be true. Got snarked at - which primarily just convinces me that I would have been better off leaving that question unanswered. As for why? I've now provided you with the mechanism to go educate yourself and consider it, which I think is pretty decent for a random, unsourced and non-specific conjecture from a mysterious and unknown other. – Fomite Apr 17 '15 at 3:46
  • On many SE sits,people propose incorrect things and want proof that they are wrong if challenged. How people handle this is up to them. If I've given a reasonable answer and I'm uninterested in doing more, I just ignore comments. If the answer is good, the votes will speak. – anongoodnurse Jul 3 '15 at 16:08
  • @Zaralynda - If you don't understand an answer, you can ask for clarification. But please remember to be nice, which goes a long way in getting someone to continue engaging with you. – anongoodnurse Jul 3 '15 at 16:15
5

I suspect that you have some mechanistic idea about how the immune responses to influenza and Dengue differ to preclude the notion of exposure to one strain of influenza causing a more severe reaction to another, a la Dengue. You could first provide some basic overview about the immunology of the influenza vaccination, citing a review article or a textbook or whatever. Then explain the idea behind how dengue does what it does (review, textbook, whatever). I imagine it wouldn’t be all that difficult to show that these are not logically compatible. No direct empirical data and no ‘proof’ (if there ever was such a thing in medicine) but we could all learn something about how somebody who understands these bugs thinks about the idea.

If I could think of a good purple skin mechanism, I’d volunteer to try it for beta blockers.* I can’t, but you do seem to have a good starting place for thinking about your question with the Dengue contrast.


*Actually, there my approach would probably be different. Since there are gadzillions of subjects who have been given beta blockers in a controlled fashion, I would find a couple big cardiology trials comparing beta blocker to placebo and point out that "purple skin" is not in their table of reported adverse events. Obviously, the study wasn't powered for that outcome, and it's by no means 'proof.' Still, if these are the most relevant available data, I think it's reasonable to present them. It seems to me that for most questions there are either theoretical (as with the virus question) or tangentially relevant (as with beta blockers) data that can be brought to bear in a way that is both reasonable and edifying despite being imperfect.

  • While probably the correct way to go, my problem is this - for many random conjecture related questions, where the original mechanism isn't even discussed (how does your friend propose this happens?), you're asking volunteers to provide a review of the immunology of influenza vaccination, the corresponding immunology of Dengue fever infection, and how the two don't actually mesh nicely together. That will, admittedly, result in very good answers even when the true heart of the answer is "This is just not a thing" - when those answers occur. – Fomite Apr 15 '15 at 23:01
  • 1
    But I will tell you that for myself, that provides a very heavy disincentive to write answers, the end result not being "we all learn something about how somebody who understands these bugs thinks of the idea", even in an under-sourced form, but the exact opposite, as experts walk away entirely, or redirect their effort to more experts-talking-to-experts sites where they will actually see some benefits (CrossValidated, TCS, Math, etc.) – Fomite Apr 15 '15 at 23:03
  • TL;DR: Is our community standard actually intended to be "The good is the enemy of the perfect"? – Fomite Apr 15 '15 at 23:04
  • Re. comment #2: I’ve spent ~10 years learning, practicing, and teaching medicine. On Health.SE, I have an idea about the answer to many of the questions, including this one: “this is just not a thing.” On the other hand, I couldn’t have come up with this Dengue contrast and, while I get the idea from from the tidbit you’ve given us, I could really learn something from an explanation. Health is broad, and this is specialist stuff (at least to those of us who don’t practice in an area where it’s prevalent.) To me that’s not undermining the expert-to-expert idea but supporting it. – Susan Apr 16 '15 at 10:44
  • Re. #3: Huh? If there is a study on the topic, is (1- β = 0.8) “perfect”? Or in the rare instance something is designed to ‘prove' the negative, is α = 0.05 “perfect”? We settle with these all the time. If no direct evidence is available, our numbers would be worse still, granted, but we work with people-->data, not pure math, and none of our data are perfect. We work with what we have. @Fomite – Susan Apr 16 '15 at 10:54
  • The root question with #3 isn't "if there's a study on the topic" but "If I can give an answer that is 'This is not a thing', plus some supporting information that builds up to a justification, should I not post that because it's not also fully cited?" Perfect in this case = fully cited answer, Good = an answer that is not everything we might want in a prototypical answer, but still correct. – Fomite Apr 16 '15 at 14:44
  • 2
    @fomite I think I got lost there. I’m certainly happy with ‘this is not a thing’ + some support. It’s a matter of how much ‘some’ is, and since we’re talking about a situation where there is no direct evidence, whatever you think explains it conceptually to the OP seems adequate to me. I’m just saying that these are not impossible to answer. I’m not so good with aphorisms, let alone their converses. ;-) – Susan Apr 16 '15 at 14:56
8

"Can taking beta-blockers turn your skin purple?"

I like this example.

I think an answer saying "to my knowledge this was never proven and this doesn't appear on the list of known side-effects of beta-blockers (source)" would be totally acceptable... until someone comes with a better answer should a specific study exist.

The rationale behind my thought is that it would be hard to close such questions as they might be completely valid and useful to others. And I see no problem in them remaining unanswered for a while, and why not featured.

-3

Pubmed is your friend.

It finds two papers for the search: beta blocker AND purple. The papers don't seem relevant from their titles, but changing "purple" to a broader term that references skin pigmentation would yield even more hits that would need to be read.

Always be humble in the face of nature. Any negative answer should refer to relevant Pubmed searches.

  • 2
    No offense, but telling someone who mentions Pubmed in their question that "Pubmed is their friend" is a little counter-productive. Not even sure what "Always be humble in the face of nature" even means and, well, the 'always refer to relevant Pubmed searches' is even an incomplete suggestion - if we're suggesting that, then your failure to also refer to the relevant Google Scholar and EMBASE searches might just indicate that you didn't look hard enough, as coverage between those three is non-overlapping. – Fomite Jul 2 '15 at 22:42

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .