Talk:Dementia with Lewy bodies

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Feedback from Cas Liber[edit]

  • Material from paras 1 and 3 of REM sleep behavior disorder subsection should be amalgamated. Also looks a bit contradictory at first glance (though ultimately isn't) Cas Liber (talk · contribs) 11:18, 14 April 2018 (UTC)
  • Not thrilled about the antipsychotic recommendations as they themselves are based on a 15-year old paper that predates the release of aripiprazole, which is (I suspect) what most people would start with if needing an antipsychotic, but whatever. I might even ditch para 2, which could be confusing. Starting with "antipsychotic medications that should be used with great caution,..." after pointing out their grave dangers might lead some readers to think these are the ones to use first rather than last.... Cas Liber (talk · contribs) 11:43, 14 April 2018 (UTC)
  • Thanks, Cas, will work on these ... but I have some 2015 and 17 papers on treatment that discuss antipsychotics, so will see what they say about aripiprazole. SandyGeorgia (Talk) 13:50, 14 April 2018 (UTC)
  • I am not finding any indication on aripriprazole. Case studies saying good, case studies saying bad. I cannot access this, but the google scholar abstract is: "Pharmacotherapy in Dementia with Lewy Bodies, M Ikeda - Dementia with Lewy Bodies, 2017 - Springer … Dementia with Lewy bodies (DLB) is the second most common type of senile dementia following Alzheimer's disease (AD) [1]. The … Despite an attractive in vitro profile (a partial dopamine agonist), aripiprazole can induce serious extrapyramidal side effects such as parkinsonism … " This review does not say anything more about aripripazole than clozapine or quetiapine. This is not promising either. SandyGeorgia (Talk) 14:46, 14 April 2018 (UTC)
  • None of this surprises me. Thing is, at every understaffed and underskilled nursing home around the world...we all know what drugs staff will be using on unruly patients with dementia, don't we...? Cas Liber (talk · contribs) 19:32, 14 April 2018 (UTC)
  • Yes ... some of the sources did go in to how to very cautiously approach ... but getting in to that kind of detail feels to HOWTO. Will see what Tryptofish thinks? SandyGeorgia (Talk) 19:35, 14 April 2018 (UTC)
  • @Tryptofish: what do you think? On the one hand, I feel that education about antipsychotic use in DLB is reinforced in every journal review, but Cas has a point that they way it is phrased now "might lead some readers to think these are the ones to use first rather than last". SandyGeorgia (Talk) 15:14, 14 April 2018 (UTC)

    Antipsychotic medications that should be used with great caution, if at all, for people with DLB include chlorpromazine, haloperidol, olanzapine, risperidone, and injectable antipsychotics.

  • Rephrase and dequote The genetics are "vastly understudied"
  • Done, but I also rejigged the order a bit, to get risk factors together, and to explain that the genetics are understudied before explaining the issues. [1] OK? SandyGeorgia (Talk) 15:22, 14 April 2018 (UTC)
  • The precise mechanisms contributing to DLB are not well understood, and a matter of some dispute. - the material following does not give an indication of a difference of opinion that a statement like this suggests. Needs some explanation.

Rest of it looks good. Cas Liber (talk · contribs) 12:06, 14 April 2018 (UTC)

@Casliber: @Tryptofish: re this edit, our article says thioridazine is still available as a generic; would it be OK to reinstate this? SandyGeorgia (Talk) 13:57, 14 April 2018 (UTC)
If you insist I am not too bothered really. One almost never sees it any more though Cas Liber (talk · contribs) 19:26, 14 April 2018 (UTC)
What I think may really be going on here is that first generation neuroleptics (some of which are still used quite a bit and others have largely been supplanted) are contraindicated in DLB, whereas the newer, second generation atypical antipsychotic drugs are (1) increasingly the drugs used for psychosis in general, and (2) may in several cases be safer in DLB because of their (basically) lesser relative specificity for dopamine receptors. It may be better to dispense with listing the names of these drugs entirely, and wording it in terms of drug classes. Casliber, what do you think about that? --Tryptofish (talk) 22:21, 14 April 2018 (UTC)
I encountered somewhere wording about those that act on D2 receptors being the worst. But if taking out the list entirely is the easier way to go, that works. I am not finding any indication that newer antipsychotics are necessarily better, and the news that come out this week about deaths from pimavanserin gives me pause that we should be very careful here ... I am particularly worried, as Cas says, that by including any list at all, we may give the impression others are safe. SandyGeorgia (Talk) 22:54, 14 April 2018 (UTC)
It's possible that it would be better to word it in terms of which receptors they act at, rather than broadly in terms of new and old, but that has the potential to get very complicated, and of course I don't want to do any WP:OR. I'm inclined to go with whatever Casliber would recommend. --Tryptofish (talk) 22:02, 15 April 2018 (UTC)
I think it is a vexed area - for instance, studies not showing quetiapine is effective, yet reviewers recommending it because of low side effects and ignoring the fact that it had not been shown to work. Furthermore the two groups (typicals and atypicals) are more heterogenous wthin themselves than certain member drugs are to each other, rendering the distinction spurious in my opinion. Hence, my concern over lists in general. In vitro affinity doesn't necessarily translate to in vivo EPSE, a la aripiprazole - but if the data is lacking I am loth to make assumptions. Anyway, those are my thoughts. Cas Liber (talk · contribs) 13:38, 16 April 2018 (UTC)


@Cas, Tryptofish, and Anthony: thanks! So, I propose we:

  • Drop this entirely:

    Antipsychotic medications that should be used with great caution, if at all, for people with DLB include chlorpromazine, haloperidol, olanzapine, risperidone, and injectable antipsychotics.

  • Change in the lead,:

Now:Antipsychotics, even for hallucinations, should be avoided because people with DLB are sensitive to them, and their use can result in death.

Change to: Antipsychotics, even for hallucinations, are usually avoided because people with DLB are sensitive to them, and their use can result in death.

Separately, I will look for the statement about antipsychotics that act on D2 receptors being the worst offenders, and once I find it, we can discuss adding something. SandyGeorgia (Talk) 13:44, 16 April 2018 (UTC)

  • Yes, I strongly support that lead change, per WP:NOTHOWTO.
  • I also agree about removing the drug list. Given the source below about D2 receptors, I think something like

Antipsychotic medications with D2 dopamine receptor blocking properties are used only with great caution.

would work. --Tryptofish (talk) 22:49, 16 April 2018 (UTC)
List of antipsychotics avoided removed. SandyGeorgia (Talk) 04:07, 19 April 2018 (UTC)
And replaced with statement about antipsychotics that act on D2 receptors, per Tousi 2017. SandyGeorgia (Talk) 04:11, 19 April 2018 (UTC)
Looks good, thanks. --Tryptofish (talk) 18:13, 19 April 2018 (UTC)

Other antipsychotic feedback[edit]

Even if not usable, what do others think about including it in External links? It is quite good. SandyGeorgia (Talk) 18:14, 16 April 2018 (UTC)

SandyGeorgia (Talk) 14:14, 16 April 2018 (UTC)

In my opinion, all three of those are reliable sources, and I think the Tousi one is a good source to say that about D2 receptor block. --Tryptofish (talk) 22:42, 16 April 2018 (UTC)
The letter poses some problems - ideally you'd want the information in a Review Article...and then if it is in a Review begs the question about why we need the link to the letter. The mededicus link appears to be a Review that we could use yes. Regarding Tousi, I'd dispute his findings (maybe he lives in a nice part of the world with well-funded and managed nursing homes) but what he says is in the article now. My more realistic wording is "antipsychotic use is widely discouraged (with varying degrees of success)" - but that is only my opinion....and his is the source. Cas Liber (talk · contribs) 03:33, 17 April 2018 (UTC)
NB: If antipsychotics were only rarely prescribed all these folks wouldn't be making such a big deal of how bad they are!! Cas Liber (talk · contribs) 03:33, 17 April 2018 (UTC)
Cas, I don't think Tousi meant to say it's no longer a problem; the context of Tousi's sentence is that severe antipsychotic sensitivity is no longer listed as a Core feature, rather a Supporting feature, because some progress had been made on that front, with better recognition about antipsychotics that work on D2 receptors. But others are killing patients as well. Unfortunately, it is the letter that makes that point best (Emergency Department physicians' lack of DLB knowledge kills patients). I will keep looking through reviews. What are opinions about using the letter in External links? SandyGeorgia (Talk) 11:42, 17 April 2018 (UTC)
Added Emergency Dept info to External links. SandyGeorgia (Talk) 04:19, 19 April 2018 (UTC)
About Tousi, I think we can say that antipsychotics are typically avoided, without getting into how well or poorly the problem has been solved to date. Also, the Tousi source appears to be a sort of editorial, rather than a primary source report of findings, so it seems to me to fit with the MEDRS meaning of a review, even though it isn't named as such (sort of what is sometimes called a "mini-review"). But if it looks like the preponderance of sources are saying that most antipsychotics should not be used, and there is no identifiable pattern that identifies the ones that seem safer, then we should simply refer to antipsychotics in general, without getting into the names of individual drugs. In that case, I would say that the sentence about "Antipsychotic medications that should be used with great caution, if at all, for people with..." should be removed. If we have enough sourcing to say that antipsychotics are avoided, except for [specific drug names] that appear to be safe, then we can say that. But no need to give examples of the ones that are avoided. --Tryptofish (talk) 23:17, 17 April 2018 (UTC)
  • Boot 2015 seems to be still the best bet if we want to add more text-- go down to the "Hallucinations and delusions" section. SandyGeorgia (Talk) 12:19, 17 April 2018 (UTC)
  • And McKeith Consensus (emphasis mine):

    The use of antipsychotics for the acute management of substantial behavioral disturbance, delusions, or visual hallucinations comes with attendant mortality risks in patients with dementia, and particularly in the case of DLB they should be avoided whenever possible, given the increased risk of a serious sensitivity reaction. Low-dose quetiapine may be relatively safer than other antipsychotics and is widely used, but a small placebo-controlled clinical trial in DLB was negative. There is a positive evidence base for clozapine in PD psychosis, but efficacy and tolerability in DLB have not been established.

    SandyGeorgia (Talk) 12:28, 17 April 2018 (UTC)

Please pardon typos:

p. 215: When using antipsychotics in DLB or PDD, the likely balance of risks vs. benefits requires very careful consideration.
p. 216: DLB patients can be exquisitely sensitive to antipsychotic agents for delusions and hallucinations and develop life-threatening sensitivity reactions ...
p. 228: These patients are particularly sensitive to developing extrapyramidal symptoms (EPS) and also to the potentially fatal complication of neuroleptic sensitivity, which affects ~50% of DLB patients.
p. 229: Analyses of pooled data from RCTs indicate that the use of antipsychotics in older individuals with dementia is associated with an increased risk of cardiovascular disease and mortality. Although clozapine is useful in treating PD psychosis, very few studies have been conducted in patients with DLB or PDD. ... Olanzapine appears to be poorly tolerated in a considerable number of patients, even at low dosages ... Risperidone has been associated with a high risk of neuroleptic malignant syndrome ... aripiprazole can induce serious extrapyramidal side effects such as parkinsonism and tardive dyskinesia. Quetiapine reduces reduces psychiatric manifestations of DLB wihtout casuing neuroleptic sensitivity or increasing EPS ... no siginificant differences in the primary outcome of efficacy ... may be an attractive candidate at this point.

SandyGeorgia (Talk) 12:48, 17 April 2018 (UTC)

"Should" and similar language[edit]

What I was trying to get at when I was talking about contraindication above is that we should not be using prescriptive language (i.e., sentences that use words like "can, could, may, might, shall, should, will, would, must, have to, has to, and ought to" to convey obligation or permission) in our articles; those are normative statements and the only exception where it's potentially appropriate for inclusion in an article is when those statements are directly attributed to a source. This article should only contain positive statements because encyclopedia articles are supposed to contain factual statements about the article topic.

Prescriptive statements that pertain to clinical practice also constitute medical advice, so the inclusion of these contradicts our own WP:Medical disclaimer when we write them in Wikipedia's voice (i.e., when we don't explicitly attribute a statement like "XYZ should ..." to the source making the claim within the same sentence). The attribution of that form of language, however, makes it clear that Wikipedia itself is not providing the medical advice in the article text; rather, Wikipedia is simply repeating the advice from the attributed source. Using prescriptive language in a sentence and citing a source from which it originates without in-text attribution to that source isn't sufficient for its attribution for the same reason that this practice is insufficient for the attribution of direct quotes: it's not explicit.

So, just to point out a few instances of where unattributed prescriptive language is used in this article (NB: compare the statements below to the second paragraph of Mild cognitive impairment#Treatment):

  • A prior history of violence or injury should be inquired about, as they increase the likelihood of future injurious dream enactment behaviors.
  • A multidisciplinary approach should go beyond early and accurate diagnosis to educating and supporting the caregivers.
  • To lessen the risk of fractures in individuals with DLB who may be at risk for falls, bone mineral density screening and testing of vitamin D levels can be done,[1] and caregivers should be educated on the importance of preventing falls.
  • Driving ability should be assessed regularly, and medical alert bracelets or notices about medication sensitivity "can be life-saving".
  • Visual hallucinations associated with DLB create a particular burden on caregivers, so they should be educated on how to distract or change the subject when confronted with hallucinations rather than arguing over the reality of the hallucination.
  • Caregivers and people with the condition should be educated about the increased risks of antipsychotics for people with DLB; if evaluation or treatment in an emergency room is needed, they may be able to explain that, in the words of Gomperts "patients with DLB are essentially 'allergic' to haloperidol and other neuroleptics".
  • Bed partners may be advised to sleep in another room, and firearms should not be kept in the bedroom.

In the bulleted article text above:

  • Bold text → prescriptive statements
  • Underlined text → uses "may", but expresses possibility, so not prescriptive

The problem with the bolded text can be fixed by attributing the source or by simply rephrasing the statement so that it's a factual/positive one. For example, "antipsychotics shouldn't be used in people with dementia" is a normative statement, but "antipsychotics are contraindicated for dementia" is a positive/factual statement; only the latter one is encyclopedic.

In any event, that's all I have to say on this at the moment. I hope I've made it clear as to why statements like this are problematic. It's not absolutely necessary to fix the language in this article since so many others also have this issue; however, it does need to be addressed if this article is going to be nominated at FAC. Seppi333 (Insert ) 04:39, 18 April 2018 (UTC)

Thank you for sharing this, Seppi333. I am fairly certain that every use of should and such is backed by the cited source, but naturally, once the bigger issues in the text are ironed out, I will be combing through and doublechecking every instance.
With respect to the idea that each of these word usages must be explicitly attributed inline to a source, I do not see any place in the (helpful) list above that you cite any Wikipedia policy or guideline page that requires same. My responsibility is to make sure I am faithfully and accurately reflecting what high-quality, recent, secondary reviews say.
I am not sure if you are aware that the proposed policy ATTRIBUTE was specifically rejected by a community-wide RFC in 2007 (Wikipedia:Attribution/Poll)? Among the reasons it was rejected was that it would dilute Wikipedia's core policies and lead to the kind of cumbersome editing practice you describe above ... that is, we would be required to inline attribute dozens of clearly cited and accurate statements in this article. That ATT was rejected perhaps explains why we have so many articles that do just as this example does, as you mention.
As one example that I was reviewing just yesterday relative to WIAFA, the lead of dengue fever contains the cited but unattributed statement: "Nonsteroidal anti-inflammatory drugs (NSAIDs) such as ibuprofen should not be used. While the vaccine section of that article has some very rough prose that might have a problem at FAC or FAR, this is a perfectly acceptable use of "should". [1]
I am also aware that there has been an attempt to get attribution in via a back door in a guideline, and should the issue come up, I am just the person to deal with that.
Another thing is that (historically, at least, although with such a dropoff in participation, it may have changed-- my sense is that FAC just isn't happening anymore), FAC delegates/coordinators have read through hundreds of debates that involve personal preferences and guidelines, and are well aware of the distinction between policy, guideline and personal preference. Objections must be actionable, and when the nominator can show that an objection is grounded not in policy, but in personal preference, the FAC coordinators are empowered to overlook it. That is, if the article comes to FAC with cited but unattributed instances of "should", you would need to produce a policy page that describes the practice you advocate.
I am most appreciative of the reminder to go back and check every instance of prescriptive wording versus the sources, but if you are still determined to force attribution into an article where policy does not require it,[2] I suggest that you would have to revisit a community-wide redo of the ATT RFC, which caused quite an issue at its last outing. SandyGeorgia (Talk) 15:05, 18 April 2018 (UTC)


  1. ^ The source says: "Do not give acetylsalicylic acid (aspirin), ibuprofen or other non-steroidal anti-inflammatory agents (NSAIDs) as these drugs may aggravate gastritis or bleeding." It also says: "Massive bleeding may occur without prolonged shock in instances when acetylsalicylic acid (aspirin), ibuprofen or corticosteroids have been taken."
I agree with Seppi about this. For me, this is different than general prescriptive language because we are dealing with medical content, and that is more important than FAC history. I do not think it's sufficient to say something like "X should be avoided.[1]" Something like that must be: "According to Humperdinck, X should be avoided.[1]" And generally, I would do that only when it's really the best way to say it. Most of the time, it is better to change it to a non-prescriptive construction. And that is easy to do, so there is no reason to argue the community history about it. If there's a tough passage, just ask me and I'll be very happy to figure it out. --Tryptofish (talk) 17:39, 18 April 2018 (UTC)
Tryptofish, convince me based on the dengue fever example (which uses the word should four or five times, to Parkinson's disease's seven-- both articles recently subject to external peer review). The source says Do not give ... (ibuprofen) and massive bleeding may occur ... How is it wrong to say that sources back the wording "Nonsteroidal anti-inflammatory drugs (NSAIDs) such as ibuprofen should not be used".
Paraphrasing medical content is already hard enough; why should we introduce a need for wording change that is not supported by Wikipedia policy, anywhere? SandyGeorgia (Talk) 20:10, 18 April 2018 (UTC)
Wikipedia:Manual of Style/Words to watch does not mention this issue. I think a discussion about normative/positive writing for various article topic kinds might be useful, but I'm not aware of any guideline. There is a problem here in that medical care is not physics with hard facts that can be tested, nor is it economics where they haven't worked out that care with writing style doesn't help it be any less of a pseudo-science than astrology :-) To become a treatment really does require value judgement. There is evidence, of course, but it may be sorely lacking, and must also be combined with subjective opinion such as bearable side effects, acceptable cost, and appropriate risk/reward. We aren't writing an academic piece in which we advance arguments to make a case, where certain writing rules help guide to make stronger arguments.
As soon as we list agreed treatments then the reader is advised. Wikipedia can claim it does not offer medical advice all it likes, but it does, and there's not really any way around that if we are at all selective about what we claim to be a treatment, etc. The phrases "indicated for" and "contraindicated in" are merely medical jargon for "should be used to treat" and "should be avoided when". A difficulty with in-text attribution is that the reader may wonder if this is the only body/person holding that opinion, or wonder why this body/person is considered an authority on the matter, rather than someone else. If the advice is undisputed and/or widely offered, then attribution may be misleading. I disagree with some of Seppi's examples -- a couple of them are just statements of fact. I'd prefer to avoid "should" if possible, but concerned we end up with lots of passive statements, which can be boring as well as incongruous wrt serious concerns or needs. If Tryptofish wants to try to rephrase some of the "shoulds/etc" in the article then I'd be interested to see the result. -- Colin°Talk 20:32, 18 April 2018 (UTC)
None of those statement I cited are factual; each of them expressed a readily apparent value judgement. Seppi333 (Insert ) 20:33, 25 April 2018 (UTC)
I just did this. It is not that hard. I have been pinged for it in bio articles with cultivation/aviculture sections Cas Liber (talk · contribs) 20:40, 18 April 2018 (UTC)
I wish we would/could finish the writing before honing in on issues like this :)  :)
What do people think of the Causes, Pathophysiology, and the History sections? Are they complete? They are the best I can do so far, do I need to dig deeper? I'd really like to see how we stand on content and then be able to move on to combing through every use of the word should (not that I would mind if someone else looked at them, but no, I will not generally use inline attribution when it is not called for by any guideline or policy when it was explicitly rejected by a community-wide RFC). And I agree with Colin that we fool ourselves if we think the language we switch to is any less "medical advice" than the word "should" is. Wikipedia gives medical advice, period, and should have a visible disclaimer on every article. SandyGeorgia (Talk) 21:04, 18 April 2018 (UTC)
Cas Liber, I don't think that change fixed anything regarding the concern of Seppi. The phrase "is strongly discouraged" is still a value judgement, still a normative statement, and begs the question "by who?". Really these writing concerns are generally for making statements about the economy or advancing philosophical arguments. I do have to laugh at some of the economics websites that give examples of "positive statements" that make claims of being "testable". The best we can do is not be bloody obvious about offering medical advice, and "should" is an easy tell. -- Colin°Talk 21:17, 18 April 2018 (UTC)
Casliber strongly discouraged is no different than should be avoided. Seppi is saying we need to add inline attribution, even though all sources support either choice of wording, and there is no requirement anywhere on Wikipedia to provide such intext attribution (in fact, it was expressly rejected by community-wide RFC). This is not a trivial matter; there is not a medical FA on the books which meets the personal preference set forth by Seppi333, and extending editing beyond what policy requires should not be done without careful consideration. By Seppi's definition, we have no medical FAs. And we fool ourselves in a non-trivial manner about the level of medical advice Wikipedia most clearly does give if we think swapping out instances of "may" or "should" will solve the problem. In dengue, you take an NSAID, you can die. It is OK to say that. With DLB, you take an antipsychotic, you can die, too. Who are we fooling with these word choices? The sources support the facts. SandyGeorgia (Talk) 22:02, 18 April 2018 (UTC)
I didn't see this comment earlier: By Seppi's definition, we have no medical FAs. I should point out that neither of the medical FAs I've written use prescriptive language that isn't attributed in the same sentence to the cited source. Seppi333 (Insert ) 20:27, 25 April 2018 (UTC)
So much discussion over something that is so easy to fix! I am basing what I am saying on WP:NOT, which is a fundamental policy. From WP:NOT#Wikipedia is not a manual, guidebook, textbook, or scientific journal, first numbered point: Describing to the reader how people or things use or do something is encyclopedic; instructing the reader in the imperative mood about how to use or do something is not. I've done this: [3]. You're welcome. --Tryptofish (talk) 23:12, 18 April 2018 (UTC)
@SandyGeorgia: WP:NOTHOWTO - I will vehemently oppose promotion of this article at FAC if every single instance of these normative medical statements isn't removed prior to its nomination. Seppi333 (Insert ) 00:33, 19 April 2018 (UTC)
So, would you like to hear that your vehemence has me shaking in my boots, unimpressed, or merely aware of what a wonderfully collaborative editor you are? I am still hoping for a rational discussion of the very appropriate (and fully supported by the sources) use of should in the lead of dengue fever. SandyGeorgia (Talk) 02:55, 19 April 2018 (UTC)
I'm such a wonderful collaborator, aren't I? Face-smile.svg To be frank, I don't really care about how you feel about it because the article's promotion isn't up to you. It really surprised me that your immediate reaction to my mentioning the use of unencyclopedic language in this article was to argue with me about it though. Given that you were a FAC coordinator at one point, why would you of all people want to lower the quality threshold of featured articles? Seppi333 (Insert ) 02:58, 19 April 2018 (UTC)
@Colin: and @SandyGeorgia:, I feel that "strongly discouraged" has a subtly different emphasis to "should". However, I personally don't see it as a deal-breaker for FA-hood. If at the end of the FAC day Seppi is opposing on that and everyone else is not, then I suspect the coordinators will take that into account. Agree with Sandy that other areas need processing first and leaving this for the time being. Cas Liber (talk · contribs) 00:43, 19 April 2018 (UTC)
PS: I take most secondary sources proposing X as a treatment are by definition acting on general information as they are review articles. There are some exceptions but they can be argued at the time (eg antidepressants in bipolar depression...sigh) Cas Liber (talk · contribs) 00:45, 19 April 2018 (UTC)
@SandyGeorgia: Just to be clear, the purpose of providing in-text attribution is to change a normative statement by Wikipedia into a positive statement about a normative statement made by the source. "XYZ should..." is a normative statement, but "A says that XYZ should..." is a positive statement. Encyclopedias contain facts, not opinions. The latter one, despite expressing an opinion, is a factual assertion about an entity expressing an opinion. In other words, it's factual because it's falsifiable (i.e., either that entity did or did not make that assertion). Seppi333 (Insert ) 00:57, 19 April 2018 (UTC)
Does anyone have a problem with [4]? --Tryptofish (talk) 00:58, 19 April 2018 (UTC)
Perfectly fine with me. Seppi333 (Insert ) 01:00, 19 April 2018 (UTC)
Yes, there is a serious problem, Tryptofish, which is that the statement is not true and there is probably not a source that supports it (as Cas points out, the problem is that antipsychotics are still prescribed all the time). What is rarely prescribed anymore are specifically the D2 receptor antipsychotics, but that is not the point to made in the lead. The point to be made in the lead is about the severe sensitivity to all antipsychotics, of any type. This is why there is a section above to help us sort out how to address this particular issue con calma, rather than as a reaction to bright yellow highlighting on the page. The statement that we now have in the article is inaccurate. Could we please go back to calmly discussing text, as collaborators, rather than with vehemance? It usually gives much better results. SandyGeorgia (Talk) 03:33, 19 April 2018 (UTC)
Hi Sandy, I think that when I made this change: [5], it was an improvement. And when you made this further edit: [6], that was a significant further improvement, and I fully agree with it. I hope you understand that I am entirely calm and happy to collaborate. Whoever may be vehement, it is not me. --Tryptofish (talk) 17:38, 19 April 2018 (UTC)
  • [7], [8], [9], and [10]. I think that covers everything on the list. --Tryptofish (talk) 01:37, 19 April 2018 (UTC)
    • Keep in mind that the only sentences that I included in that list were the ones which used the term should and didn't mention antipsychotics, since the antipsychotic sentences were already under discussion. Searching for related terms in the list I provided (i.e., the use of can, could, may, might, shall, should, will, would, must, have to, has to, and ought to" to convey obligation or permission) will need to be done to ensure that all of the unattributed prescriptive statements are removed. Most clinical practice guidelines I've read make frequent use of the terms that I've bolded to prescribe a specific approach to clinical care. The words "can" and "may" are often used to convey possibility as opposed to permission, so one needs to consider the usage of these terms when they're encountered. I can look later and supply a list of sentences that use prescriptive language as I did in the list above, but I'm not inclined to do it right now. Seppi333 (Insert ) 01:48, 19 April 2018 (UTC)
I'm happy to do these, and I'd actually prefer a pointer to which section instead of a list. Of course, it would be great if other editors would make these kinds of edits too. --Tryptofish (talk) 02:09, 19 April 2018 (UTC)
I'll indicate where they are by tomorrow then. The only other issue that stood out to me when I went through this article pertains to the alignment of some of the images (see below). Seppi333 (Insert ) 02:58, 19 April 2018 (UTC)
I'm going to wait until the article is more fully written before I address the wording in various sentences. I'd end up doing more work than necessary otherwise. Seppi333 (Insert ) 20:27, 25 April 2018 (UTC)
Thanks, Tryptofish. There is one unintended change of meaning at:[11]
  • Regular assessment of driving ability, and medical alert bracelets or notices about medication sensitivity "can be life-saving".
By combining the two sentences, we ended up with regular assessment of driving ability can be life saving, which is not what the source says ... it was two different thoughts. I will come back and figure out how to remove the one attribution that snuck in, because it's what all sources say, and we don't want to leave the impression it is only the opinion of one reviewer. Thanks again, SandyGeorgia (Talk) 02:46, 19 April 2018 (UTC)
Does this wording suffice: [12]? Seppi333 (Insert ) 03:42, 19 April 2018 (UTC)
Tryptofish, I revised to this. The sentence was already quoting Boot directly, so quoting those three more words and attributing the entire sentence to him should solve both problems without introducing too much more verbiage. I am at a loss for a good way to rephrase, "Regular driving assessment should be conducted." Thanks again, SandyGeorgia (Talk) 04:28, 19 April 2018 (UTC)
That's fine, I think attributing it to Boot is entirely OK, no problem. --Tryptofish (talk) 17:48, 19 April 2018 (UTC)

In the interest of removing instances of should, inaccuracies (now three) are being introduced. We now have:

Regular assessment of driving ability is a common practice.

Seppi, do you have a source for this? The cited source says no such thing. Would you mind, please, not making text changes without consulting sources? SandyGeorgia (Talk) 03:44, 19 April 2018 (UTC)

As you'll notice, I reverted the edit after I committed it because my intention was to link to a diff with the use of the proposed language in the article given that I didn't WP:V-check what I wrote. I'm on a crappy wifi right now. I'm not going to not use example diffs if my intention is to merely revert my editing and link to the diff. Seppi333 (Insert ) 03:48, 19 April 2018 (UTC)
OK. And I missed your revert in my reply; I follow every edit to make sure we don't get any unintended changes. There is no need to rush here; slow and steady wins the race. We have had a very collaborative, and deliberative pace until today. I suggest we go back to that mode of operation. We have a discussion above about how to address the antipsychotic issue. We haven't even heard from most editors, and suddenly changes back and forth, resulting in three successive errors being introduced ... please strive for stability and collaboration-- there is no hurry. SandyGeorgia (Talk) 03:56, 19 April 2018 (UTC)
If I understand correctly, [13] is the current version, and that's fine with me. --Tryptofish (talk) 17:52, 19 April 2018 (UTC)

Ok, I think Seppi's argument about normative/positive language is a total red herring and has no place here. I think he's got confused that statements that provide instruction that are loaded with an implied reason for following them, are both normative and "providing instruction". We want to avoid the latter, or at least being rather obvious about it. I'm no grammar expert, but I think he's got these two aspects of language mixed up. I would currently support reverting the attempts to avoid "should" because most that I have looked at change the meaning, make claims our sources don't, and are pretty awful passive English.

Tryptofish's example of "The use of antipsychotics, even for hallucinations, is strongly discouraged" to "Antipsychotics are rarely prescribed" and the joint change of "Driving ability should be assessed regularly "Regular assessment of driving ability is a common practice" both change meaning and also make claims our sources don't. Guys it is really really important you don't do this. Most of the sources of this article are available to read, though of course Sandy has the benefit of access to multiple sources on each point, which makes it much easier to put words into our own language. I saw this problem when we had students mass editing medical articles, who were told: find a journal article and add a factoid to the wiki article. It was very very hard for them to avoid plagiarising while at the same time writing something that was correct and supported by the source -- they didn't really understand the subject well enough, of have access to enough literature, to do a good job of finding original text.

Copyediting is surprisingly hard, and despite Tryptofish's claim that rewording these is easy, he hasn't managed it. What you have done is the difference between "You should avoid obesity, keeping your waist measurement below half your height" to "People are generally slim". You've assumed that the advice is actually followed. And at the same time, removed any imperative for following the advice, so it becomes a meaningless factoid carrying no weight.

The change "A prior history of violence or injury should be inquired about, as they increase the likelihood of future injurious dream enactment behaviors" to "A prior history of violence or injury may increase the likelihood of future injurious dream enactment behaviors." Is wrong, Tryptofish, this is not the "Prognosis" section, but the "Clinical history and testing" section. So we need to describe what history (i.e. the medical practice of enquiring about the patient's life, course of disease, not the abstract concept of previous events) the doctors will take and what tests they are advised to carry out. Similarly in the Caregiving section, we need to describe what advice is given to patients.

A statement "keeping firearms in the bedroom is considered to be dangerous" is a bland statement that most people outside of North America would respond with "Well, doh!" and wonder why it belongs in this article rather than as a banner advert shown to all readers logged in from the US ;-). The former text gave the reader the understanding that this was advice given to patients and caregivers. So can we find a way to keep the fact, and these are facts, that this advice is given, because we're writing a "Treatment - caregiving" section, and avoid making irrelevant obvious statements like "guns are dangerous".

I agree we should try to avoid explicitly instructional language. But it is hard, and harder than I think some editors here think. Attributing is one possibility but can have issues I raised above. Perhaps there is an authority we can cite for some of the claims, but I would like to avoid name dropping random physicians. Is there a patient charity who have a set of recommendations? Perhaps we can introduce a set of statements to make them clear they are the sort of advice offered to patients/caregivers (or earlier, the advice given to doctors when taking history, or when considering prescribing drugs). Because advice is a fact, is encyclopaedic: this is the advice, the recommendations. When we remove the advice, we make those sections meaningless. -- Colin°Talk 08:12, 19 April 2018 (UTC)

That was not helpful. You are free to make edits that correct any edits that I have made. --Tryptofish (talk) 17:54, 19 April 2018 (UTC)
The type of language that is used in this article to "provide instruction", as you put it, is normative, but not all normative statements are prescriptive (NB: a command like "do this" is a form of instruction an instructional statement - that's neither fact/positive nor opinion/normative; commands clearly do not belong in an encyclopedia either though). The only reason I used the term "normative" instead of a more specific one is that the opposite of a normative statement is a positive statement - i.e., a fact - which is the only type of statement that belongs in an encyclopedia. FWIW, I prefer to just say "According to the authors of [number] reviews(s)," when attributing a claim, as opposed to name drop, but there's obviously other ways to do this.
I'm well aware of how tedious it can be to rephrase prescriptive medical statements from reviews and medical guidelines because I've had to do it in other articles on many previous occasions. Sometimes it's simple, sometimes not so much - depends upon the statement and context. Seppi333 (Insert ) 08:56, 19 April 2018 (UTC)
The simplest way to make any normative statement - prescriptive or otherwise - encyclopedic is to attribute the statement to the source; as I said above, adding attribution to these yields a factual statement. Rephrasing these statements, as opposed to attributing them, isn't actually necessary and can be detrimental in some cases (e.g., your examples). Seppi333 (Insert ) 09:38, 19 April 2018 (UTC)
Colin, the last 24 hours of editing were most unfortunate. I have been at clinic for three days, and just wanted last night to keep up with the errors. Once I am home, I will step back through the whole thing, re-check everything, and re-do anything if necessary (I hadn't noticed some of what you mentioned, and that will be easier from a real computer at home).
Seppi, I might suggest keeping in mind that the goal is to have the best information available on the Internet about DLB-- not a bronze star. Your brightly colored threat to vehemently oppose at FAC for this article if your personal preferences are not instated was at best immature and uncollaborative, and at worst malicious and battleground-ish. I hope you understand that to a FAC coordinator, an oppose is an oppose is an oppose, and it matters not how strongly, brightly or vehemently you place it. It will be evaluated on its merit, of which yours has little, and FAC coordinators are well empowered to disregard invalid opposes. So please try to realize that such behavior will only make things more unpleasant for other editors, with little change to the article.
We can progress faster towards an excellent article by discussing and collaborating, rather than rushing to introduce changes that result in errors, as happened over the last 24 hours. If the article doesn't get a star, I really don't care-- there's some satisfaction in still holding the record at FAC, and knowing what an FA is and is not whether or not it holds a little icon in the corner. Let's please have a return to collaborative discussion. In spite of Tryptofish's statements that all of this discussion was over a simple matter that was easy to solve, deciding how to best handle the language about the fact that antipsychotics kill people with DLB is not a trivial matter, and why we have to get the language right. I hope the last 24 hours serve to exemplify why we should be discussing, not vehemently forcing our personal preferences into an article by threats against a meaningless bronze star.
Because real people read our articles. SandyGeorgia (Talk) 13:05, 19 April 2018 (UTC)
Here's another one (fatal gun incident). SandyGeorgia (Talk) 14:36, 19 April 2018 (UTC)
Again, I am not being vehement, and I urge editors not to personalize these things. I'm sorry you've had a tough couple of days, really. I am trying to help. --Tryptofish (talk) 17:59, 19 April 2018 (UTC)
Yes, you have been helpful! Thanks so much, @Tryptofish: Just that yesterday, with me away at clinic for two nights over, there was a bit of us all stumbling over each other :) I will catch up this weekend. I appreciate that you move forward and make edits, because it seems that we are stalled at times. Thanks again. SandyGeorgia (Talk) 18:15, 19 April 2018 (UTC)
OK then, good! --Tryptofish (talk) 18:18, 19 April 2018 (UTC)

And, GOOD news and BAD news. The good news is, I stopped and picked up three books. The bad news is, there is almost nothing useful in them, so I wasted my money. The best news is that this article is already so far beyond what is available to readers at their bookstore, that I was pretty amazed at how bad the books were relative to the ground we have covered. I do have some things to tweak as I get to them... can do so much better on Causes and Pathophysiology, and now know where to look, but no help on the "should" issues from the books. That we have been able to update to the newest diagnostic criterion and pull in the most recent journal reviews, puts us well ahead of anything I can find in any books! Cool beans, SandyGeorgia (Talk) 18:16, 19 April 2018 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────@SandyGeorgia: Sigh. Look, I understand your frustration Sandy. It's often difficult to write high-quality medical articles for Wikipedia. I'm not trying to make things unnecessarily difficult for you. I'm simply trying to ensure that a high-quality article is written. My assertion that it's necessary to attribute opinions - and advice in particular - to a source isn't a view that's limited only to me. Not writing an opinion in Wikipedia's voice (i.e., an unattributed opinion - what I've been calling an "unattributed normative/prescriptive statement") comes directly from WP:WikiVoice (1st bullet); for the explanatory supplement to the MOS which elaborates on that point, see WP:ASSERT. Not writing advice in Wikipedia articles comes from WP:NOTADVICE. That said, these policies do permit one to write advice in Wikipedia articles provided that it's attributed to a source; doing so expresses a fact about an opinion. I'm not going to argue these points with you. If you want to ignore me and these guidelines, feel free to do so. I don't think that would be a wise decision. Seppi333 (Insert ) 18:43, 19 April 2018 (UTC)

@Colin: I wasn't entirely awake last night so I didn't really address your initial point. The reason it's not a red herring to be discussing positive (i.e., factual) and normative (i.e., opinionated) statements here is that the MOS repeats the following points on several pages: (1) opinions should not be expressed as facts and (2) opinions should be attributed. The first point is stated in several MOS subpages. The second point is actually stated in the parent guideline, WP:MOS. Both points are made together in WP:WikiVoice. Seppi333 (Insert ) 18:43, 19 April 2018 (UTC)

Seppi333 thank you for striking and removing the brightly colored vehemence-- the gesture is most appreciated! I had "heard" you early on, but felt we needed to work towards building consensus about how to best deal with certain terminology and important issues in this article, and the vehement opposition just wasn't needed for us to all work together. I am most grateful that you rethought and struck.
Where we differ is on how you are distinguishing fact from opinion, as explained at WP:WikiVoice. "NSAIDs should not be given to people with dengue fever" is not opinion. It's a scientific fact, expressed by multiple MEDRS sources. "Antipsychotics should be avoided in people with DLB" is not opinion; it's again, a fact backed by scientific evidence in multiple MEDRS sources. So while I feel that you are overinterpreting policy and overapplying guidelines, that we have a difference is no longer relevant. What is relevant is that we seek a high quality article, we need to figure out how to best phrase these things, and when we are all less tired, we will get there. We might not get there as fast as we'd all like, but slow and steady wins the race! (I am so tired today after three days of hospital, and had not even realized I was wiped out last night, so it wasn't a good editing night all round!) SandyGeorgia (Talk) 19:34, 19 April 2018 (UTC)
I find it useful to look at the "should"-etc. issue in terms of whether a reader (particularly an unsophisticated one) might be led by the choice of words into thinking that Wikipedia is telling them that they, the reader, should do something that affects their own health care. That's not so much a matter of opinion-versus-fact, as about making sure that we are not implying that what we say is medical advice from editors. --Tryptofish (talk) 20:16, 19 April 2018 (UTC)

I've got limited time at the moment. Tryptofish, the reason I'm not reverting/revising your edits is I'm not a reverty sort of editor. I would rather we all came to an understanding. As Sandy notes, the aim here is to provide high quality and accurate information for the reader, not to get a star or to satisfy one person's idiosyncratic grammar opinions. It is far more important that if there is consensus advice towards doctors/patients/carers then we accurately describe that, than worry so much about how we word it, that we end up losing the advice and, worse, making false statements, unsupported by our sources.

I maintain the whole normative thing is a red herring. We are not writing an essay that advances a case, where the presentation of facts and objective claims (true or false or dubious) rather than the author's opinions and values is what the teacher is looking for. I think we'd make better progress by looking to ensure Wikipedia's voice is not offering advice or opinions but that it is clear this is the advice and opinion of the profession and scientists. If Seppi wants to keep arguing about it, then open an RFC (unrelated to this article) and get MOS to be changed. Or perhaps Sandy knows some grammar/language experts.

The problem with statements like "opinions must be attributed" is that it really isn't always black and white. As I noted earlier, the fact that a drug is indicated for treating a condition is a medical opinion -- it is not totally based on facts and comes loaded with value judgements by various bodies. All we've done is hide "should be used to treat" inside an "indicated for". By using medical jargon, we're signalling to the reader that some authority has formed and published this opinion, but haven't in fact named that authority. That's a clever trick because it no longer looks so much like "Wikipedia's opinion" but more "Expert opinion". And we've also avoided using obvious weasel words. It is possibly harmful to name just one authority, but other times where expert opinion varies (e.g. UK use the drug but US does not) it may be useful to name one authority. The point is that by adding in-text attribution, you are changing the meaning -- you are signalling something about the authority you quote. If, in fact, the advice is totally consensus and widespread, then by attributing one body, you are signalling a very wrong thing to the reader. Opinions are on a spectrum. I think the best approach is that we document medical advice, clearly signalling with our language that it is official medical advice, rather than what editors here think readers should do, but being careful not to wrongly signal that this advice is possibly contentious and held only by one author/body. -- Colin°Talk 08:03, 20 April 2018 (UTC)

Colin, thank you for discussing these points in a non-condescending way. I'm happy to discuss these issues with you, and I think that you will find that I am not unintelligent. And I'm also entirely agreeable to coming to an understanding in talk. I thought that I was making helpful edits, so I went ahead and made them. Nor did I say that you should have simply reverted me. Instead, I pointed out that you are free to make edits that would further revise my edits, just as Sandy did here: [14].
First of all, I want to agree with you that it can be a problem to attribute something to a single source when it is actually widespread consensus. If you look back at the discussion, it was Seppi, not me, who argued for attribution; all I did was not actively disagree with him. (All I really did in that regard was: [15], following [16].)
Instead, as I said just above your comment here, my concern is about avoiding giving readers the impression that we (editors) are giving medical advice or instructions, per WP:NOTHOWTO. If we tell readers "your condition should be treated with Drug A" but their physician prescribed Drug B, it would be awful if Wikipedia caused the reader to disregard the physician's opinion. That's what I was doing here: [17] and [18] – no attributions, and in multiple cases further improved by Sandy. But I want to be very clear: we are here to provide information, not advice. WP:MOS is a guideline, but WP:NOT is fundamental policy. --Tryptofish (talk) 21:13, 20 April 2018 (UTC)
Tryptofish, "condescending" is when an editor puzzles aloud why he is surrounded by idiots making a meal of a triviality, boasts of being able to easily fix it, and concludes with a patronising "Your welcome.". And when the argument descends to wikilawyering over "fundamental policy" vs "guideline" then, really, who is being condescending? Sandy and I weren't born on wiki yesterday. Could we try to discuss this without claims that it is simple, or statements that your concerns are based on "fundamental policy" when in fact, all our concerns (accuracy, sourcing, weight etc) are also "fundamental policy". It is a balance of priorities and some compromise might be required. Wiki fundamentalism and vehement opposition aren't consensus-seeking.
We all agree we should avoid being obviously in Wikipedia's voice when advice is presented. Writing "is favoured" or "is beneficial" is offering an opinion in Wikipedia's voice. It doesn't jump out perhaps as much as "should" but it is there all the same. Perhaps that's enough to satisfy some, but this just shows how hard it really is to follow some rule that all opinion must be attributed. Can we agree that if authorities recommend advising patients & carers to remove firearms from their bedroom, that that "advice" is a "fact" that we should consider presenting in some way. It is not at all the same to say "keeping firearms in the bedroom is considered to be dangerous" because that is unsourced and does not present the same information. It isn't our job to deconstruct the advice back to what we guess were the fundamentals behind the advice, and offer them in the hope the reader might form the appropriate conclusion. Similar for the antipsychotics -- we can't just present the facts that they may have adverse effects if actually the warning advice itself is a notable fact with sufficient weight to warrant repeating here. -- Colin°Talk 15:43, 25 April 2018 (UTC)
I've had enough of this garbage. I am now taking this page off of my watchlist. Do not ping me or ask me to come back. --Tryptofish (talk) 00:03, 26 April 2018 (UTC)
It's really not that hard to write an FA-quality article which is devoid of unattributed medical advice. Yes, it can difficult to figure out how to word these statements in some cases, but it's not like it takes hours of deliberation to decide on how to accurately convey the idea while complying with content policies/guidelines. For the antipsychotics, one could just state the black-box warning, attributing it to the USFDA. There really isn't a better drug authority to cite for safety information than that agency (NB: there are comparable authorities for drug safety). Seppi333 (Insert ) 20:27, 25 April 2018 (UTC)
Putting the changes here, so I will remember to look at them when I have time to spread all the sources out and revisit. SandyGeorgia (Talk) 12:06, 20 April 2018 (UTC)
@Colin: I think we'd make better progress by looking to ensure Wikipedia's voice is not offering advice or opinions but that it is clear this is the advice and opinion of the profession and scientists. Given that this is exactly what I've been asserting all along, I don't understand why you've been disagreeing with me. Seppi333 (Insert ) 21:20, 20 April 2018 (UTC)
@Sandy: I'm glad to see that the article is going in the right direction now. I still haven't had a chance to thoroughly check for other prescriptive statements that use the verbs I previously listed, but I will do that sometime in the near future. FWIW, I'd be happy to vehemently support (718smiley.svg) at FAC once the text is cleared up and I've had a chance to do a check against certain FA criteria (1b, 1c, 2 - the MOS in general, 2a, 2b, and 3; 1b will require a lot of time on my part).
Also, since CNS physiology and pathophysiology are subjects I find interesting and often write about on WP, I'm willing to help write the pathophysiology section for this article if you still intend to expand it. Seppi333 (Insert ) 18:49, 21 April 2018 (UTC)
I am reworking Pathophysiology now in sandbox, because we had no mention of amyloid and tau ... I will ping here when done, for review. (The article was always going in the right direction :) I needed time to finish and review prose. ) I actually could have gotten there much faster without the agida ! SandyGeorgia (Talk) 19:04, 21 April 2018 (UTC)

Proposed solution[edit]

I am averse to attributing "should" text to a specific review because the only kind of should text in this article can be sourced to multiple secondary reviews-- it is generally accepted advice, such as the dengue fever information that NSAIDs should not be given, which is well sourced.

In searching for a way to attribute such text, I have looked into using one of Colin's ideas. What do others think of this proposal?

  1. We have a secondary review that recommends the Lewy Body Dementia Association (LBDA).
  2. We have multiple secondary reviews that mention the subjects that we are discussing above, so attributing them to any one review seems misleading to our readers (they are information that is broadly supported).
  3. So how about if we switch some of these instances to the voice of the LBDA, as long as and only when the information can also be cited to a MEDRS secondary review. That is (example),
    The LBDA says Joe Bloe should not eat worms.(cited to LBDA)(cited also to secondary review about eating worms).

Would that work for everyone? SandyGeorgia (Talk) 14:55, 19 April 2018 (UTC)

I'm fine with attributing views in the form of "the LBDA says". --Tryptofish (talk) 18:16, 19 April 2018 (UTC)


The idea is to incorporate the secondary review, while also making it clear this is a broad recommendation, not confined to one reviewer. So, I wanted to work in both the secondary review, and the broader issue. How is this?[19]

  • Driving ability may be impaired early in DLB because of visual hallucinations, motor issues related to parkinsonism, and fluctuations in cognitive ability; Ian McKeith (DLB researcher) and the Lewy Body Dementia Association advise that assessment of driving ability should take place early,[42] and Boot adds that assessment should be conducted regularly.[13]

Will work on others as I get to them-- got sidetracked sorting out the Robin Williams mess. SandyGeorgia (Talk) 15:11, 20 April 2018 (UTC)

Sample 2:

Retained the secondary review, but worked in broader advice from LBDA:[20]

  • Visual hallucinations associated with DLB create a particular burden on caregivers;[41] caregivers should be educated[13] on how to distract or change the subject when confronted with hallucinations rather than arguing over the reality of the hallucination, according to the Lewy Body Dementia Association.[42]

SandyGeorgia (Talk) 15:36, 20 April 2018 (UTC)

Sample 3:

Caregiver education on meds made more general:[21] (It's not just beneficial it's serious!) With these last two changes, I am satisfied that I am caught up with all the edits from a few days ago.

  • According to the Lewy Body Dementia Association, caregivers and people with the condition should be educated about medications used to treat DLB, and the increased risks of antipsychotics for people with DLB;[43] if evaluation or treatment in an emergency room is needed, they may be able to explain that, in the words of Gomperts "patients with DLB are essentially 'allergic' to haloperidol and other neuroleptics".[16]
  • Prompt evaluation and treatment of RBD is indicated when a prior history of violence or injury is present as it may increase the likelihood of future injurious dream enactment behaviors.[5]

SandyGeorgia (Talk) 16:01, 20 April 2018 (UTC)

I'm still uncomfortable with directly attributing LBDA, McKeith or Boot, as a general solution. Concerned we have replaced one problem with another, while also making the text more verbose. Also concerned that some of the previous edits made (see above) had replaced what is merely a concern about tone/language with text that is actually factually wrong and unsourced. Unfortunately I'm not feeling inspired with a great solution nor finding a whole lot of time IRL to read/write the text. Will keep thinking about it. -- Colin°Talk 15:43, 25 April 2018 (UTC)

Above, Tryptofish writes "If we tell readers "your condition should be treated with Drug A" but their physician prescribed Drug B, it would be awful if Wikipedia caused the reader to disregard the physician's opinion." The thing is that we do do this. While we generally try to avoid saying "should", we do offer the medical opinion about what drugs are indicated for treating a disease, and "indicated for" is far far stronger than just saying "X is used to treat" -- it is saying that X is one of the treatments that should be used. Furthermore, sometimes we say "X is the first line treatment for Y", and again this is just an opinion, unattributed in-text, and may well cause the reader to question their physician's choice. The fact that some treatments are first-line, some are last-restort and some have been superseded is notable information that often has weight for us to include in an article. We wouldn't argue for censoring those consensus opinions just because a reader may ignore their doctor's wiser choice for that individual. Nor do we have folk insist that all prescribing information is in-text attributed. I don't think the reader should have to learn specifically that McKeith, Boot or Gomperts think something, and a bit false to attribute one author among many in a paper.

It is really important that trying to satisfy one policy, we don't just end up breaking another. Can we agree on that? This newspaper article says "Thousands of people in the UK with a commonly misdiagnosed form of dementia are at increased risk of injury or death because they are being prescribed potentially harmful antipsychotic medication, experts have warned.". There's a world of a difference between finding a way to repeat the medical advice on treatments, diagnostic tests, preventative measures and contraindications, and making a claim that this is happening already. If we must wikilaywer about it, and we need to find a way to balance competing concerns, then our Wikipedia:Core content policies win.

FWIW, I'd be tempted to just drop the firearm advice. It is a uniquely American concern, and will simply puzzle readers from most other countries. That section lists several safety ideas/recommendations, but they seem to be more a relatively arbitrary list of "things to try". What is the message? Our sources say "Improving safety is the mainstay of management of RBD", "Non-pharmacological measures are prudent in every patient", "All patients with RBD should be counseled about bedroom safety principles to prevent injury or serious consequences". I note the words "mainstay", "prudent in every patient" and "all patients... should be counseled". So the experts feel that every patient & carer needs counselling about the need to improve bedroom safety. That's a fact, widely noted so has WEIGHT, and should be mentioned in the section called "Management / Caregiving". It is not equivalent to just list safety measures in a passive voice. I also don't see the need to attribute the fact "bed partners may be advised to sleep in another room" to "St Louis and Boeve". Perhaps we can think of the passive hints-at-expert-opinion language that "indicated" and "first line treatment for" gives us, to avoid the advice/opinion being in Wikipedia's voice, but also avoid clumsy and arbitrary attribution to journal-article authors. -- Colin°Talk 18:33, 25 April 2018 (UTC)

I haven't gotten to this yet, Colin ... SandyGeorgia (Talk) 20:31, 26 April 2018 (UTC)

Above Seppi333 writes "I should point out that neither of the medical FAs I've written use prescriptive language that isn't attributed in the same sentence to the cited source":

  • "amphetamine ... is also contraindicated in people currently experiencing advanced arteriosclerosis (hardening of the arteries), glaucoma (increased eye pressure), hyperthyroidism (excessive production of thyroid hormone), or moderate to severe hypertension" The source for this is the patient information leaflets by Arbor Pharmaceuticals LLC and Shire US Inc. As noted above "is contraindicated in" is just medical jargon for "should not be taken by" and is medical opinion. Would the reader benefit from attributing in-text? It would, as I note above, give a false impression that those are the opinions just of Arbor Pharmaceuticals, and Shire, rather than being more widely held. We'd certainly be wary of that if we started attributing the indications sections to pharmaceutical company patient information leaflets.
  • "People who have experienced allergic reactions to other stimulants in the past or who are taking monoamine oxidase inhibitors (MAOIs) are advised not to take amphetamine" this is the same as "should not take amphetamine" but written in the passive voice. They "are advised" by who? Wikipedia? Experts?, Big Pharma? The Government?
  • "These agencies also state that anyone with anorexia nervosa, bipolar disorder, depression, hypertension, liver or kidney problems, mania, psychosis, Raynaud's phenomenon, seizures, thyroid problems, tics, or Tourette syndrome should monitor their symptoms while taking amphetamine" I can see you jumping through hoops to in-text attribute that advice, but the result is awful.
  • "Drug abuse" the article uses the term "abuse" several times. This is a concept without formal definition, and which there is considerable disagreement about what it means. Merely using the term is a value judgement and opinion about what consitiutes "abuse".
  • "Addiction is a serious risk with heavy recreational amphetamine use but is unlikely to arise from typical long-term medical use at therapeutic doses." Just a note that the quotes from your sources do not support the claim in the latter half.
  • "According to one review, an optimal dosing regimen is to administer it in one 1 gram dose, three times a day, since this ensures elevated plasma concentrations of HMB throughout the day" The source for this PMID 27106402 appears to be just a narrative review. In other words, the opinions of Thomas Brioche, currently a postdoctoral researcher on the “muscle remodeling and signaling” team at Montpellier University; Allan F. Pagano, a Ph.D. student at the University of Montpellier on the “muscle remodeling and signaling” team; Guillaume Py, Ph.D., a cosupervisor of the “muscle remodeling and signaling” team at Montpellier University; and Angèle Chopard, a Professor at the University of Montpellier, who teaches physiology and anatomy in the Faculty of Sports Sciences, and a scientific member of the “muscle remodeling and signaling” team. So here we have opinion about optimal dosing that isn't even medical opinion (these people are just scientists, students, lecturers), isn't consensus opinion, isn't backed up by anything approaching a MEDRS recommended RCT, and in-text attributed to "one review". Seppi333, while a systematic review can have conclusions that may appear to be objective, a "narrative review" can't have an opinion, any more than "according to some book, ..." or "a newspaper article I read said that ..." Are you really going to in-text attribute Brioche, Pagano, Guillaume and Chopard? You can't attribute the "muscle remodeling and signaling" team because we don't know if their opinions represent the rest of the team. And I'm sure the University of Montpellier would be reluctant to take credit for supposed medical advice by a bunch of PhD students and their sports science lecturer.

Mostly your medical FAs have been about pharmaceutical compounds, which can be described objectively and factually. As soon as we start talking about people, humans, things get messy and less clear. What is abuse? What is optimal? What is safe? All just opinions. I'm not raising these points so you must go fix those articles, but just to show that it is actually hard. And we need to apply some common sense rather than dogma. -- Colin°Talk 07:41, 26 April 2018 (UTC)

"People who have experienced allergic reactions to other stimulants in the past ... " SandyGeorgia (Talk) 12:35, 26 April 2018 (UTC)

Merge proposal[edit]

As a reward for two long days of work, and spinning off Notable cases as discussed here, please see the merge proposal at Talk:Lewy body dementia. That does me in for the day. Done for a while, would appreciate others going through prose ! SandyGeorgia (Talk) 01:03, 21 April 2018 (UTC)

New To Do List[edit]

If there is anything above I cannot archive in a few days, please speak up! SandyGeorgia (Talk) 01:28, 21 April 2018 (UTC)
  • Deal with merge proposal at Talk:Lewy body dementia
  • Work in concommitant AD/DLB including tau pathology, etc. [22]
  • Decide on expert review
  • Sort Tryptofish concern about memory relative to AD, and AD relative to DLB, here.
  • Does @LeadSongDog: still have concerns (lost track of list)?
  • Decide on imaging abbreviations, here.
  • Can anyone find any more info for Causes?
  • Consult Looie496, Graham Beards, Eric Corbett for copyedit.
  • Deal with antipsychotic issue in lead
  • Over or under Wikilinking ?
  • Did all the delinking I (think) I intend to do, intentionally linking some difficult things more than once. SandyGeorgia (Talk) 00:27, 22 April 2018 (UTC)
  • Trim ? Suggestions?
  • Found 200 redundant words already :/ SandyGeorgia (Talk) 04:34, 21 April 2018 (UTC)
  • AFTER all of that is done, ping in WT:MED for a look.

SandyGeorgia (Talk) 01:29, 21 April 2018 (UTC)

I saw that you un-hatted the discussion about pimavanserin. Is there an unresolved issue about that? --Tryptofish (talk) 19:51, 21 April 2018 (UTC)
No, it is good. I had hatted things off (temporarily) so I could see what I still needed to do. For archiving, I wanted to unhat, so that we don't invalidate future archive searches. I unhatted everything above so I can hopefully archive it all, and we can begin the next pass of truly fine tuning. SandyGeorgia (Talk) 19:55, 21 April 2018 (UTC)


I thought we were almost there. But when I noticed that this text from the original lead was not adequately paraphrased from the source[23] and needed to be rephrased, I started checking that source, and realized that since March 24 (and earlier), [24][25] the article has used sources for Lewy body dementia, not Dementia with Lewy bodies. That is, some of the text applies to Parkinson's disease dementia but not dementia with Lewy bodies. I should have noticed sooner :( :( So, now I am going back through checking every statement sourced to these NIH sources, which are all LBD, not DLB. I am finding errors, removing some things, resourcing what I can, finding some that are still OK because they are general rather than specific to DLB ... anyway, sigh ... back to the drawing board, just when I thought we were just about done. I am going to try to resource as much as I can ... I am finding so many issues that I have no confidence in any sourcing to NIH, NIA or NINDS. There are a number of things sourced to them that I have not seen in any secondary review. Do not feel comfortable approaching FAC at this stage. SandyGeorgia (Talk) 04:06, 22 April 2018 (UTC)

I think I fixed it all. The remaining text cited to NIH, NINDS and NIA sources on LBD is now contextually appropriate, and I've switched to DLB sources where specifics were needed. That was discouraging. SandyGeorgia (Talk) 07:25, 22 April 2018 (UTC)
Paraphrasing can occur but is not required as these sources are public domain as produced by the US government. This tag however should be used if not fully paraphrased {{PD-notice}} Doc James (talk · contribs · email) 01:01, 26 April 2018 (UTC)
Doc James, yes, I know ... I wasn't so worried about that other than it could set off red flags at FAC. I was more set back by realizing the sources were LBD not DLB, needing to go back and check everything, and the muse has not yet returned. I appreciate the list below and will get on it in a few days (some are intentional, I will need to explain, hoping to get another source), but for today, just too tired. (Curiously, after a bad week, I found we have no article on radiation colitis, it is a redirect to radiation proctitis, which is not the same thing.) Will address the list below by the weekend. Thanks again, SandyGeorgia (Talk) 01:35, 26 April 2018 (UTC)


  • "The autonomic nervous system is usually affected, and can result in symptoms like changes in blood pressure, heart and gastrointestinal function, and constipation"

Constipation is a change in gastrointestinal function thus this sentence reads strangely. Doc James (talk · contribs · email)

Ugh. Constipation is a very common symptom, and one that presents early and may be present before other symptoms, so my intent was to specifically highlight it. How is this?

SandyGeorgia (Talk) 03:54, 26 April 2018 (UTC)

  • "specific biomarkers, blood tests, neuropsychological tests, medical imaging, and polysomnography."

The biomarkers are "polysomnography" and certain types of medical imaging.

Neuropsychological testing is how dementia is diagnosed and is summed up within "based on symptoms". It is a core part of diagnosis not a supportive feature.

How is this?
SandyGeorgia (Talk) 04:00, 26 April 2018 (UTC)
  • Would summarize "Average survival 5.5–7.7 years from disease onset" to "Average survival 6.5 years from disease onset". This is an overview and taking the middle point of the range is perfectly appropriate in the infobox.
This is a place holder; I am waiting for a new source. The numbers are all over the map, and I need to see what another source says. SandyGeorgia (Talk) 03:54, 26 April 2018 (UTC)
Got the third source, it was not helpful in sorting the differences, went back to simpler NINDS data. SandyGeorgia (Talk) 03:17, 27 April 2018 (UTC)
  • These "Other frequent symptoms include visual hallucinations; marked fluctuations in attention or alertness; and sometimes slowness of movement, trouble walking, or rigidity." are also core features. They appear to be given less weight than REM. Why the "sometimes" for the motor symptoms?
Yes, the wording is awkward. Here is what I wanted to convey (but didn't).
First, only one of the cardinal features of parkinsonism need be present, not all three. Second, the symptoms of parkinsonism are usually much less severe than in Parkinson's. Third, the symptoms of parkinsonism are not always present early in the course of the disease (which I see I read in a source but failed to add to the article ... tomorrow).
On the other hand, the other three core symptoms (RBD, visual hallucinations, and fluctuating cognition) are either present early on, or may be prodomal. I was attempting to draw a distinction between those three and parkinsonism, and "sometimes" isn't cutting it. I will need to find the quote in the source about parkinsonism not always being noticed early on. I was concerned that the word "core" would mislead the reader to think all of those signs of parkinsonism had to be present and noticeable early on, as the others are. Not sure yet how to fix this. Ideas ? SandyGeorgia (Talk) 04:15, 26 April 2018 (UTC)
Cannot figure out how to solve this in terms simple enough for the lead, so took out "sometimes". SandyGeorgia (Talk) 03:20, 27 April 2018 (UTC)

Doc James (talk · contribs · email) 01:01, 26 April 2018 (UTC)

@Doc James:, I have been through these if you would like to take a fresh look at the lead. SandyGeorgia (Talk) 03:20, 27 April 2018 (UTC)

Preclinical DLB[edit]

Regarding these edits by AJVincelli (talk · contribs · logs):

Thank you for your contributions, AJVincelli, but there are multiple serious things wrong with your edits.

First, please have a look at WP:BRD, WP:COI and WP:MEDRS.

Bold text additions are a good thing, but when you are reverted, you should start a talk page discussion, rather than WP:EDITWAR to reinstate your edit. (I would have started the talk page discussion myself, but was at clinic investigating a serious side effect of treatment my husband is having, and was editing from my phone-- now I am home.)

Second, that you have, in essence, edit warred to install primary research of your mother's in an article is a problem with Conflict of Interest.

Third, the text is not adequately sourced.

But none of that is the most troubling of what you have done. I can FULLY relate to your frustration about missing information on Wikipedia (I created radiation colitis yesterday, before I learned today of the more serious Clostridium difficile infection), but you do not even appear to have read the article before plopping an anecdotal list into it, because as of two weeks ago, this article DOES mention prodomal and preclinical symptoms in DLB. In fact, I have gone to great lengths to separate out what symptoms are likely to be seen early on (see the talk page section just above this one). All of that information is available in secondary sources, and it is not necessary to add your mother's primary sources. Not only that, your content is mostly a list form of the symptoms already discussed in the article, lending again to the impression that you did not read the article. Please do not reinstate your text again. If there is consensus here, we can add more content about the early signs, but I feel that has been covered exhaustively.

I apologize for my impatient tone; it is related to my day at clinic and not at you, but please don't do this again. SandyGeorgia (Talk) 19:31, 26 April 2018 (UTC)

@AJVincelli: just to keep everything in one place, you did the same thing at REM sleep behavior disorder, not noticing apparently that DLB is already linked in that article, and your edit added nothing new. SandyGeorgia (Talk) 19:33, 26 April 2018 (UTC)

Hi SandyGeorgia, I'm new to this Talk thing, so I'm hoping that this message reaches you. Thanks for directing me here. Sorry also about the "edit war," I thought that was the only way to reply. It said something about leaving a comment when I clicked "undo." I'll avoid "edit wars" in the future, and post here instead.
It appears that I am WAY out of my depth with these Wiki edits! I have been meaning to add my list of DLB preclinical symptoms to Wikipedia for over a year now, and I finally have an afternoon available, but apparently I have no idea what I'm getting myself into.
Basically, since my stepfather's death and the family's horrible experience with dementia, I thought that it would be incredibly helpful to include a list of examples of preclinical symptoms on the DLB page, and (more importantly) link to it from the following pages:
These were all pages that I had visited back in the early days of my stepfather's disease, but which didn't direct me to the DLB page. I really, really wish they had! It was another year before he was officially diagnosed. Silly us.
I'm hoping that others may benefit from my family's experience, so that they aren't searching or ignorant about reality of the disease afflicting their loved one. But what to do when there are little to no published data on the subject? Wikipedia is on the cutting edge, and I suspect that the scientific journals are far behind (if this info ever will actually be published, which I doubt).
I appreciate how Wikipedia is a well-supported and scientifically-sound resource for professional as well as personal purposes, and I wouldn't want to impact the quality of this excellent article with superfluous or irrelevant information. But, I do hope that it will educate others. I firmly believe that key, critical information is missing from this article.
Is there a way to include the information without over-claiming it? I thought that the word "anecdotally" would be a strong flag, as well as "may include but not limited to," but apparently those phrases are not sufficient.
So I've given the raw info that I had gathered, and I've been attempting to cite some sources, but I'm a novice with Wiki editing. My availability is limited, and my time on this is short. What can I do? Any suggestions you have are much appreciated!
P.S. -- I'm sorry to hear about your husband! I hope he is okay.AJVincelli (talk) 20:04, 26 April 2018 (UTC)
Hi, AJ; thank you for the well wishes. I apologize for having been short after a hard day. Yes, I see this message because I have this page watchlisted (I have been working night and day to improve the page for over a month ;) If you could familiarize yourself with WP:MEDMOS and WP:MEDRS, and then give this article a thorough read, we could better discuss any content that might be missing. What you should note is that the article is exclusively sourced to secondary reviews (no primary studies). There is published data on most of what your text mentioned, and I have been digging, digging and digging to get it all in. Please let us know exactly what you think is missing, but also please spend some time examining whether items you want to add are truly missing (I am not seeing that from your list), and whether they can be cited to MEDRS-quality sources. LBDA is not a MEDRS source. Executive function, REM sleep behavior disorder, restless legs are covered. Obviously, dementia is mentioned. I have never seen post-concussion syndrome in a review, but can look for it. Best regards, SandyGeorgia (Talk) 20:14, 26 April 2018 (UTC)

the new content[edit]

added here:

Preclinical symptoms

In the preclinical stage of DLB, the person is usually in otherwise good or excellent physical health. Rather, only the brain is affected, and thus the disease is subtle[1] and may be very difficult to observe directly.[2] The disease is not able to be diagnosed at this stage,[3] though the Mini-Mental State Examination (MMSE) may be used to assess the level and progression of cognitive impairment.

Data on the preclinical symptoms of DLB is scarce.[4] Symptom examples may include (though are not limited to) the following:

  • A rigid arm while holding hands and walking with a loved one
  • Occasionally inappropriate or irrelevant conversation contributions
  • Restless legs syndrome
  • Numbness in feet
  • Shuffling or not picking up feet when stepping up onto something elevated
  • Occasional poor decision-making skills
  • Vivid dreaming
  • Throwing oneself out of bed while sleeping (see RBD)
  • Dizziness and disorientation
  • Falling
  • Sleeping poorly, sleeping a lot, waking up tired, requiring more sleep
  • If sleep-deprived, or very stressed, or out of routine: delusions, hallucinations, paranoia, illogical though processes, disoriented, anxious
  • Increased anxiety (especially about health or money)
  • Forgetful
  • Confused
  • Difficulty following a TV program or a newspaper article
  • Difficulty performing routine tasks such as loading the dishwasher, making coffee, using the microwave, or dressing
  • Difficulty following simple directions
  • Difficulty understanding the time of day (day vs. night)
  • Increased note-taking to help remember things
  • Symptoms consistent with Post-Concussion Syndrome (may be misdiagnosed as this illness)
  • Symptoms consistent with Mild Cognitive Impairment
  • Symptoms consistent with Executive Dysfunction or Dysexecutive Syndrome


  1. ^ DelleDonne, A; Klos, KJ; Fujishiro, H; Ahmed, Z; Parisi, JE; Josephs, KA; Frigerio, R; Burnett, M; Wszolek, ZK; Uitti, RJ; Ahlskog, JE; Dickson, DW (August 2008). "Incidental Lewy body disease and preclinical Parkinson disease". Archives of neurology. 65 (8): 1074–80. doi:10.1001/archneur.65.8.1074. PMID 18695057. Retrieved 26 April 2018. 
  2. ^ Frigerio, R; Fujishiro, H; Ahn, TB; Josephs, KA; Maraganore, DM; DelleDonne, A; Parisi, JE; Klos, KJ; Boeve, BF; Dickson, DW; Ahlskog, JE (May 2011). "Incidental Lewy body disease: do some cases represent a preclinical stage of dementia with Lewy bodies?". Neurobiology of aging. 32 (5): 857–63. doi:10.1016/j.neurobiolaging.2009.05.019. PMID 19560232. 
  3. ^ Abbate, C; Trimarchi, PD; Inglese, S; Viti, N; Cantatore, A; De Agostini, L; Pirri, F; Marino, L; Bagarolo, R; Mari, D (2014). "Preclinical polymodal hallucinations for 13 years before dementia with Lewy bodies". Behavioural neurology. 2014: 694296. doi:10.1155/2014/694296. PMID 24868122. 
  4. ^ "Is It LBD or Something Else? | Lewy Body Dementia Association". Lewy Body Dementia Association. Retrieved 26 April 2018. 

-- Jytdog (talk) 19:38, 26 April 2018 (UTC)

Thank you, Jyt. So, to the additional problems with this text.
  • In the preclinical stage of DLB, the person is usually in otherwise good or excellent physical health.
  • I doubt that can be reliably sourced.
  • Rather, only the brain is affected,
  • That is not necessarily true (cardiac, gastrointestinal are affected early on, often before other symptoms are apparent)
  • and thus the disease is subtle[1] and may be very difficult to observe directly.[2] The disease is not able to be diagnosed at this stage,[3] though the Mini-Mental State Examination (MMSE) may be used to assess the level and progression of cognitive impairment.
  • As prodomal signs of DLB, we have RBD, and as signs that often appear before cognitive impairment, we have orthostatic hypotension, hallucinations, and constipation-- ALL that are likely to show up before there is any sign of cognitive impairment.
So, the reason we source to secondary sources has to do with accuracy; this text mostly is not. I note it also includes a 2008 source. The Lewy Body Dementia Association should not be used to source medical content. And finally, ALL of your sources predate the new, 2017 Fourth Consensus. The knowledge of the role of REM sleep behavior disorder in DLB is a game changer, and that IS reflected in the article (in fact, it's why I rewrote it). Older sources have to be carefully vetted to make sure information from them is still current, relative to the change in diagnostic criterion from 2017.
If there is a symptom on your list that you would like to see better discussed, please point it out and I can tell you if I have come across it in a MEDRS source. SandyGeorgia (Talk) 19:51, 26 April 2018 (UTC)
I apologize for failing to mention that I am very sorry about your stepfather; my best wishes to you and your family for a merciful outcome. SandyGeorgia (Talk) 20:03, 26 April 2018 (UTC)
@AJVincelli:, if your frustration is that the OTHER articles do not mention DLB, then the way to correct that is to find the text you want in this article, use the source from this article (because they are high quality sources), and then add text to the other articles using the sources you find here. Just know that if you copy text from one article to another, you should always indicate in edit summary "Copied from Dementia with Lewy bodies"; this is required on Wikipedia. And take care to read to make sure the text is not already included (for example, I had already added DLB in to REM sleep behavior disorder). SandyGeorgia (Talk) 20:19, 26 April 2018 (UTC)