Wikipedia talk:Don't use today's news to contradict medical sources

Latest comment: 2 years ago by WhatamIdoing in topic WP:DUE or WP:SYNTH

Shortcut

edit

This essay could use a shortcut. What do we like best? WP:MEDCONTRADICT? WP:MEDNPOV? Something else? –Novem Linguae (talk) 19:46, 23 May 2022 (UTC)Reply

I suppose that WP:MEDDUE could be usurped; the section of MEDRS it was created for hasn't existed for years now. I kind of like MEDCONTRADICT. I'm not very creative about these things. Maybe wait until someone spontaneously creates one? WhatamIdoing (talk) 02:59, 24 May 2022 (UTC)Reply
I should have checked here first but created the WP:NOTMEDNEWS shortcut for now already, —PaleoNeonate16:51, 26 June 2022 (UTC)Reply
That sounds good to me. Thanks for doing that. WhatamIdoing (talk) 18:17, 26 June 2022 (UTC)Reply

WP:DUE or WP:SYNTH

edit

While citing news sources within an article can definitely violate WP:DUE. I can't help but feel that objections have more to do with some vague form of WP:Synth than anything else. Certain medical news stories can have real political and social implications they don't really have scientific one's (yet). The problem is that these seemingly current affair statements are actually likely to be interpreted as medical ones.

For example, Donald Trump's declarations about COVID are almost certainly relevant somewhere on wikipedia. But perhaps not right next to secondary sources on the main page.

There is also a question about whether you are barring news sources from the entire article. For example, if there was a section "Media response to X" within the article would that also be WP:UNDUE? I don't necessarily have an issue with this... but wonder if these sort WP:DUE arguments, actually turns "X" into "What people with a medical degree say about X?".

I don't really care about what politicians say about medical topics. I do start caring about what economists have to say (e.g. when they say that public health figures about the costs of obesity or STI's are massively inflated), or psychologists (e.g. when they offer an alternative treatment modality to PTSD than antipsychotics), or social scientists (when they start observing abuse that takes place in psych wards, and gender and racial disparties, or diffferences between countries, and the only thing that the medical research cares about is the abuse that healthcare professionals experience). I'm a little concerned that narrow interpretations "mainstream" and broad interpretation of WP:DUE allow a pro-medic bias in social issues surrounding medicine.

Talpedia (talk) 13:20, 25 May 2022 (UTC)Reply

Well... yes and no. It's more complicated than that. Imagine that we had these sources early in an event:
  • Prof. Big in J. Imp. Science: Combining the results from six previously published papers, it's a pandemic!
  • WHO's press release: We declare that this is a pandemic.
  • J. Journalist in The News: The WHO says it's a pandemic. See our 16 stories.
  • P. Politician's social media: No, it's not, you silly sheeple! They're just trying to tell you what to do!
These sources are (in order):
  • peer-reviewed, secondary, independent (of the event)
  • self-published, (mostly) primary, non-independent (of its own decision to declare a pandemic)
  • good editorial control, (mostly) primary, independent
  • self-published, primary, non-independent (of his POV)
The WP:BALANCE section of NPOV says "when reputable sources both contradict one another and also are relatively equal in prominence..." then both views should be included equally. But here, we have a contradiction without equal prominence. Two sources that are independent, secondary, and/or non-self-published outweigh the the politician's POV. Another is the agency that we've all agreed is supposed to officially make that declaration. If we were to write "This quasi-political organization says it's a pandemic but that political group says it's not", we would not be representing a balanced view of reality. It would give undue weight to the minority POV and equal validity to views that are not equally valid.
But I can also see why you might think of SYNTH first, because that unbalanced statement additionally tends to "imply a conclusion not explicitly stated by any source". There is an unstated, implied "So nobody really knows whether it's a pandemic or not ¯\_ (ツ)_/¯" hanging at the end of it. I would personally think of MEDRS first (it's statement about "Primary sources should not be cited with intent of "debunking", contradicting, or countering conclusions made by secondary sources" is exceptionally clear and applicable), NPOV second, and SYNTH third. But it's really all of them.
On your other thoughts:
I support "Media reactions" (and "Politicians' reactions" and whatever else seems relevant) content; however, it needs a source that Wikipedia:Directly supports that content and is "appropriate to the claims made". This could be anything from a magazine article to an editorial in the newspaper to a scholarly book, but it does need to be "about" the media reaction, and not merely an example of a media reaction. (Coincidentally, I reverted an edit with that problem just yesterday.)
When biomedical researchers have different POVs than economists/psychologists/sociologists/etc., I find that this model usually works pretty well. These are all scholars; they all have the same general approach. You write something like "Alice Expert says that tobacco causes disability and shortens lives, resulting in the equivalent of zillions of dollars of damage to society through increased medical expenses and lost potential productivity. I.M. Portant says that the many premature deaths result in lower costs to pension programs, which saves the non-smokers several of those zillion dollars in cold hard cash."
The only real difficulty is when the disagreeing people aren't operating at similar levels. The obvious example is anti-pyschiatry: The Worldwide Club of Extremely Scientific Professionals declares that Thou Must Drug Them All, going through all the ceremony of scholarly journals and practice guidelines and medical textbooks, and a small number of affected people post on social media saying "This stuff is horrible. Seriously, a rational person would rather be dealing with that voice in my head." We don't give an equal voice to unequal voices. This results in charges of bias that are at least sometimes valid (e.g., excluding or minimizing the viewpoints of some indigenous groups), but it protects in other situations (e.g., Trump's promotion of dangerous and ineffective treatments). WhatamIdoing (talk) 19:07, 26 June 2022 (UTC)Reply
I suppose I'm a little afraid of the "Don't contradict"ness seeping into every section of the article, and then selectively supressing content through people engineering the article titles and scope. I'm basically trying to construct an interpretations of policies so that we could have a "history" section on the article that mentions untrue claims in a current affairs way, while also having a medical section that is exclusively MEDRS, or at least move the history section to another article.
I guess the question is how close we allow the reporting about an untrue claim to get to a related MEDRS claim, and whether this in practice pushes reporting about untrue claims off of wikipedia, and if this is what we want. I can see the argument for having an "aura" around MEDRS claims, but does that aura extend around all of wikipedia in practice (by virtue of people adding a drop of MEDRS everywhere).
You can sort of have "different levels" between psychology versus psychiatry. Approaches used in therapeutic psychology can be pretty context sensitive and more complicated than the psychopharmacological approaches, so the "theory aspects" of them basically won't stand up to scrutiny in the same way that drug based approaches can. I would argue that even CBT, the bread and butter of evidence based psychology, isn't really evidencbased medicine when it comes too its models. Though I guess you can compare effectiveness. I definitely sometimes feel there is some "slippage" going on, where the interventions people like get assessed based on effectiveness and their theory promoted to evidence, while the one's that people don't like get their theoretical assumptions attached. I'm not really sure how MEDRS interacts with this though, worthwhile ideas will probably still show up in secondary sources.
Philosophical critiques can also get caught up in this a bit. Medicine sometimes has a habit of creating "beg the question" definitions, or extending their definitions to beg the question. Insight is an example. Medics and even more so nurses like to extend to "agree with the treatment plan". There is a risk that this then gets used as a factor in choices about detention, so "you don't want to take drugs" then becomes evidence for needing to take drugs. "Insight" seems to be a big factor in involuntary treatment for anorexia for example. A broad definition of insight is procedurally useful for medics so it's only really going to get picked up by "theoretical" framings from either philosophy, psychiatrists being philosophical, sociologists which pretty much live outside of the scientific method proper.
In the anti-psychiatry case. You have a few holdout psychoanalysts whose unproven mental models give them enough rope to think that psychotic symptoms can be treated, this group then has some interaction over with psychology through dual-trained people (and at times medicine e.g. Mark Solms) as well as patient run groups like Hearing Voices Network. I think this is enough connection to academia to bring useful findings into the mainstream should they show up. Though I guess wikipedia (and journal's etc) standards might add or remove a decade of progress depending on where they draw the line. I can think of issues with more stigma around them though, or events that create more of a stigma against psychosis. Talpedia (talk) 21:18, 26 June 2022 (UTC)Reply
Wikipedia has room for genuine disputes; there's a whole article on the Beginning of pregnancy controversy. Is it possible to have "an abortion" after fertilization but before implantation? No, it isn't, but only because people decided to define "an abortion" in a particular way. Does the "medical fact" that preventing implantation is not "an abortion" tell you whether a zygote has the same moral significance as a 5 year old? No, it doesn't.
I think there is less room for disputes that are more – I'm not quite sure what the word is. Outside mainstream morality? Thinking about your insight example, if someone has "insight" to the extent of saying "My brain hates my body's appearance. If I don't eat, I'll die. If I do eat, my brain will complain. I'd rather starve myself to death than listen to my brain's complaints", then that is a level of "insight", but we (i.e., the people who aren't suffering from this problem) don't accept dying as a proportionate response. I think there is a point at which society decides that you can't have true insight if your view is extremely different from mine. It's the same thing with persistent suicidality: We really accept only fear of dying soon as an "appropriate" reason for suicide. This is illogical ("Oh, you're afraid of dying? It might be uncomfortable, you are afraid of "losing your dignity"? Well, let me do the paperwork so you can kill yourself in a government-endorsed manner"), but aside from that, anyone who shows a strong, persistent interest in not being alive, and who says things like "Suicide is an appropriate response to this trivial temporary circumstance" and doesn't say things like "I know it's just the depression talking, so let's make a safety plan" is not really considered to have insight. There is no "scientific" way to decide whether the problem is that the suicidal person doesn't have insight into their condition vs if we don't have insight into what that person is dealing with and how their response aligns with that person's values. The "outsiders" POV would be very difficult to represent appropriately anywhere in Wikipedia, because the sources are few and weak, and the POV is held by such a tiny minority.
Another subject that might be useful for exploring is chiropractic. Spinal manipulation can reduce some kinds of acute pain, but the "story" (model?) they tell about it is basically vitalism, or perhaps at this point just marketing, so you don't discover that the same procedure with the same results is available from chiropractors, osteopathic physicians, and physical therapists, and some of them don't cost as much. The story is pseudoscientific, but the story isn't what determines efficacy.
(Wikipedia sometimes surprises me with the delightful subjects editors have chosen to write about. Today, I learned that we have an article called All models are wrong.) WhatamIdoing (talk) 01:16, 27 June 2022 (UTC)Reply
To me the issue is a sort of "scientific laundering of the moral or politcal" for people's convenience. It's not necessarily that you don't want to force or encourage people with anorexia to eat food... it's that you want to do it for the correct reasons, and put the correct protections in place. Scientific framings can sort of be used to make excuses for what we do as a society. I think the abortion case is an interesting one, while spending time on the social media's you do come across medics where you feel their aim is almost "if I stop people expressing an argument then the argument is not true".
With insight, I guess there is a difference between "I don't want to be alive" and "I think what I am doing is healthy and will have no consequences". If you coerce in the latter case what you are doing is in some sense less coercive because the person is not basing their actions on a factual correct understanding of the world. The really nasty case is misdiagnosed or episodic psychosis, if "does not believe they are psychotic" becomes evidence for "are psychotic" (though it won't be that simple if it's say accompanied with an odd worldview and a psychiatrist is trying to minimize blame associated with risk to others).
While wikipedia probably can't really support a "pro-suicide through starvation for depression" perspective, it can probably support a "these diagnoses are highly subjective and seem to be used for convenience" perspective, should arguments or evidence for this exist.
With "pseudoscientific models", I guess the question is how much the just so story "leaks" outside of its use case. If you believe the just so stories of CBT are true rather than just "CBT is sometimes an effective intervention" this can start influencing social issues etc. You can see this happen historically with some gendered pseudoscience in psychoanalysis, and more recently with some of the discourse around conspiracy theories. Here, it's not necessarily bad to have sociological theories and analogies that borrow from medicine or therapeutic intervention. It's more that when you start using theories like this they should have the weight and methodologies of humanities applied to them, rather than those of science. Science can be more trusted with a lot because it's more "definitively true", but if it's not exactly science, more an interesting hypothesis, then it can't be trusted so much. Talpedia (talk) 08:45, 27 June 2022 (UTC)Reply
I don't think the "scientific laundering" is for convenience. I think it's because science has become the replacement for religion in some cultures (modern, relatively well off, individualistic). Past generations might have said that an anorexic person must eat, because else it's suicide, which is a serious sin; current generations would say that an anorexic person must eat, because else the disease will progress and the person will die, and dying from unbalanced brain chemistry is just like dying from unbalanced liver chemistry, and we need to rectify the humors.
In the "stop people expressing an argument" range, are you imagining conversations like this?
  • So if you kill the baby...
  • So if you kill the fetus...
    • It's not a fetus yet. That's scientifically incorrect.
  • So if you kill the future baby...
    • That's emotive language and presupposes the outcome.
  • So if you kill the offspring...
    • You're presupposing the outcome again.
  • So if you kill the zygote...
    • Ha, ha, it took you too long. It's not a zygote any longer.
  • So if you end the pregnancy...
    • There's no pregnancy. Pregnancy is defined by what happens to the adult female's body, not by what happens to the egg.
  • Look, just what words do you want me to use? I'm willing to use your terminology, but you have to tell me what to call this situation in which there is a genetically distinct thing that has not yet implanted in the uterus. What do you call that?
    • How do you feel about "morally insignificant scrap of totally unimportant tissue that nobody should care about"?
I don't think we would say that psychiatric diagnoses are highly subjective. I think we would instead note that there may be difficulties with Inter-rater reliability and incomplete information. (This is also true for non-psychiatric diagnoses. "Doctor, I'm just so tired all the time" could be almost anything. Do you start with thyroid tests, diet changes, diabetes screening, anemia, sleep apnea, something else?). OTOH, I thought it was the official professional stance that psychiatric labels existed for professional convenience. Once we know enough about a neurological situation to be really certain about it, it seems to stop being "psych" and start being "neuro". And once they're "neuro", it's okay to apply psychological and social interventions. You might get real resistance to recommending a strict sleep schedule or daily family dinners to a person with bipolar, but immediate acceptance for making the same recommendations to a person with dementia. WhatamIdoing (talk) 01:27, 29 June 2022 (UTC)Reply

I think it's because science has become the replacement for religion in some cultures

That's definitely an effect. Freidson talks about this a bit in Profession of Medicine and Abbott makes this argument about psychology/psychiatry vs the church in System of Profession. In the absence of a real "ought" it can be convenient to weave one out of layers of "is" interleaved with "professional state".
I'm more thinking in terms the medics making certain diagnoses themselves, though there can be a little "cocreation" going on here. An example that medics talk about is childhood colic, which sort of exists/existed so parents don't feel responsible for their crying child, and was used to justify drugging your child with alcohol. Similar arguments have been made about psychosis. I'm not convinced - but I'm not completely unconvinced either.

are you imagining conversations like this

I've seen discussion surrounding whether certain forms of emergency contraception that prevent implantation should be considered different from other forms of contraception or not, where medics seem quite... linguistics... in their arguments. Not, "this definition is the one that is most relevant when considering medical interventions" more "these are the words and the truth". Often these terminate with "why are you disagreeing with an expert about what words means", but then social media discussion aren't necessarily of the highest quality. I guess I live in a country where "brexit means brexit" was one of the most important political slogans for 2 years so might be more aware of these effects.

I don't think we would say that psychiatric diagnoses are highly subjective.

I probably exaggerated a little there! Sure, Inter-rate reliability is an objective measure accepted by psychiatrists and the literature, so deserves to be prominent. On the other hand, there are interesting results about the correlation of gender, race, wealth, etc with diagnosis and I would wager notions of responsibility and risk (a common example is you are far more likely to be diagnosed psychosis if you are black). I think to the degree that there are reliable results about this, and researchers become more able to tease apart complicated networks of causal relations these sort of results are DEU. I agree that there is probably not sufficient evidence to say diagnoses are subjective yet, but I suspect that certain psychiatric diagnoses very much are and there isn't really evidence to the contrary. As ever, this sort of understanding is just topic understanding that can be relevant for finding literature (which should then be contextualised within the entire body of literature).

psychiatric labels existed for professional convenience

I mean... there's convenience and convenience. "Each individual is unique and this diagnoses just helps us understanding the web of causality and treatment" is one thing; "I want a means to offload responsibility onto the individual and avoid risk" (e.g. personality disorder) is another.

it's okay to apply psychological and social interventions

this is probably a real effect. I suspect the label "dementia" does give people more control over you than the label "bipolar" due to perceived certainty and depndency, and that this certainty can be misplaced. In fact, people with dementia likely are more prey to the incorrect and coercive use of antipsychotics (https://www.nicswell.co.uk/health-news/antipsychotics-still-given-to-too-many-with-dementia) than those with bipolar. People with bipolar can just stop taking medication or turning up to appointments! Talpedia (talk) 14:17, 29 June 2022 (UTC)Reply
About the contraception linguistics (which, with the recent news in the US, is certainly on my mind), I think you've made an important distinction. It'd be malpractice to tell a woman, 12 hours post-conception, that this was a fine time to get drunk because she's not "technically" pregnant yet, according to the implantation definition of pregnancy. Of course, according to the way obstetricians measure pregnancy, she's already been pregnant for at least two weeks already. Thus the "no abortions after six weeks" headlines really mean "no abortions three weeks after the pregnancy starts". WhatamIdoing (talk) 20:32, 29 June 2022 (UTC)Reply