This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Archive 5 | ← | Archive 10 | Archive 11 | Archive 12 | Archive 13 | Archive 14 | Archive 15 |
I have a procedural/administrative question. I see that there is a DS notice at the RfC draft page, based on this being related to WP:MOS. It had never occurred to me that we are working under a DS situation here. Are we? -- Tryptofish ( talk) 20:20, 4 January 2020 (UTC)
There has been a fair amount of requests for clarification and interpretation occuring at my talk page (see
[1]). Out of respect for the consensus that "The question of drug pricing is remitted to a single venue"
I note the above diff and then copy over the current discussion that had been occurring there to here to be continued. Best,
Barkeep49 (
talk)
15:51, 3 January 2020 (UTC)
Content from Barkeep49's Talk Page
|
---|
The following is a closed discussion. Please do not modify it. |
Just a note to those around here, I was just asked on DocJames talk page specifically about updating. I
FYI, User:Colin/PriceEdits contains a computer-generated list of all price/cost insertion/deletion edits to 530 drug articles by any editor since 2015. It also contains my analysis of where editors have come into conflict. Plenty examples of why I note my concerns above. To take an example of the kind of "copyedit" being suggested by User:WhatamIdoing at the MEDMOS discussion, an editor changing "the wholesale price in the developing world is" to "The median buyer price according to the International Drug Price Indicator Guide was" sparked an edit war at Lactulose. -- Colin° Talk 10:40, 2 January 2020 (UTC) |
I look forward to learning about the flaw that Sandy sees with allowing the updating of information. Best, Barkeep49 ( talk) 15:51, 3 January 2020 (UTC)
I'm back; thanks for the well wishes. While I was driving, I tried to organize my thoughts to avoid my usual verbosity. It is unfortunate to clog this page with this feedback, but I understand Barkeep's reasoning for wanting to keep this here.
I would like to unpack my concerns in steps, so as not to drop a wall of text filled with my usual typos on this page, and for us to be able to calmly digest the different components.
Starting with No. 1, I felt completely betrayed yesterday. Usually I apologize for an outburst, but I hope no one feels I have something I need to apologize for, because I pushed back from the computer and hit the break button before I completely exploded. I always apologize for misunderstandings when my wording is not careful, so in that vein, I am sorry if Barkeep felt I was questioning his integrity or saying he had disrespected me or us or done something intentionally deceptive or malicious; that is not the situation.
But yesterday did not feel like this month was well spent. The ANI started on December 5, and today is January 3; we expended the entire holiday season on this, and we all typed 'til we were "blue in the face" with unfailing good faith and very little need for, as Barkeep said, people to walk things back. (I will discuss in part 4 whether some people should have been asked to walk things back, but my concern there is not aimed at Barkeep.) Those who participated here (which I will later distinguish from the MEDLEAD RFC) showed that they can address this dispute calmly and without behavioral issues, which is part of why I do not feel we are yet to the arb stage. As we have all expressed (and I believe sincerely meant), Barkeep has gone above and beyond the call of duty and has been exemplary in his treatment of all of us. Yet, through no malintent or negligence, we came yesterday to a most unsatisfactory juncture, for reasons I will unpack bit by bit here.
Basically, I feel our good faith was taken advantage of. I believe the problem yesterday happened because Barkeep has been shielded from full information of what he was walking in to, by necessity, because we are not allowed to "rehash old grievances". So, Barkeep could not have foreseen yesterday's reaction to something that, to many of us, was entirely foreseeable, because it precisely fits the pattern we have been dealing with. I'll unpack that in part 3.
When Barkeep posted to AN, I cheered him, because it has become obvious that he was working too hard. When he retracted his AN post, I worried. I find it a bit disgusting that there are probably twice as many people who participated at the ANI than have participated at the MEDLEAD RFC, and it is unconscionable that what I asked in the ANI (when I laid out my terms) that more eyes watch these events, and Barkeep closed by asking for more volunteers, he got none. I would like to revisit the ANI when I get to Part 5.
Kevin, thank you very much for stepping forward with reassuring information, but I never thought that Barkeep's conduct back-channel needed to be questioned; he has given no reason for doubts about his integrity or how he would approach other admins. My concerns had to do with him not knowing the full picture, and perhaps unwittingly falling into something he might be unaware of.
And that relates to Part 2. Barkeep, I have emailed you a description of things that have happened to me in the past on Wikipedia-- many years past-- as a female editor. To be able to move forward, I would like to ask if either you or Kevin know why the third admin on IRC has not self-identified as Kevin did. I hope you can view my question in the context of the private information I sent you, and understand that it would be reassuring to me if that admin would self-identify. If they won't, that will force me to ask the question in a less-than-desirable way. Meanwhile, I hope we'll all give a good re-read to the ANI thread, because I want to next raise some specifics about where we stand. That's all for now; Barkeep and Kevin, please let me know if it is likely that the third IRC admin will speak up. I'll continue afterwards with Part 3. Regards, SandyGeorgia ( Talk) 22:34, 3 January 2020 (UTC)
Why was updating ok with me when I'd said even adding a tag was not? In my thinking it's different because it's not adding something new and because the embargo is designed to ameliorate the conflict not preserve bad information.And Kevin (aka L235 said:
My reasoning was that the spirit of verifiability doesn't allow us to keep old bad information if newer, equally- or better- sourced information is available -- it must be updated or removed, and the closure explicitly prohibits removing it, so updating it is the only option. Additionally, updating pricing doesn't seem to involve the same dispute as to whether prices should be included or not.So it is easy to see that both Barkeep and Kevin are operating from a position of wanting to preserve the integrity and verifiability of information to our readers, while minimizing conflict. So far so good. Well intended, makes sense. But here's the problem. We have reams of evidence above that all of the drug price information we have presented to our readers in these 500+ examples is bogus. We have computer-generated analysis that shows essentially all of the prices were inserted by one editor. I have been asking for a month, and we have not yet seen one single example of these database sources being used correctly. So, questions are:
In the United States the wholesale cost is about US$13.50 a month. [3]
we used the World Health Organization DDD classification to analyse dose–response relationships. This classification implies that a dose of 25 mg of chlorthalidone is equivalent to a dose of 25 mg hydrochlorothiazide. This assumption is probably not valid. A recent cross-over trial suggest that chlortalidone is about 1.5–2 times as potent as hydrochlorothiazide with regard to antihypertensive efficacy.-- they admit using the DDD for a purpose that WHO explicitly warn against:
"DDDs do not necessarily reflect therapeutically equivalent doses of different drugs and therefore cannot be assumed to represent daily doses that produce similar treatment outcomes for all products within an ATC category". We are similarly abusing DDD for our cost-to-treat claims, made all the worse since there is no published information about what exact indication a DDD was calculated for.
A meta-analysis of trials of chlortalidone for high blood pressure found that lower doses of chlortalidone (e.g., 12.5 mg daily in ALLHAT study) had maximal blood pressure lowering effect and that higher doses did not lower it more. (Musini VM, Nazer M, Bassett K, Wright JM (May 2014). "Blood pressure-lowering efficacy of monotherapy with thiazide diuretics for primary hypertension". Cochrane Database Syst Rev (5): CD003824. doi: 10.1002/14651858.CD003824.pub2. PMID 24869750.)
We see a trend to greater benefit with 25mg, who is "we" and what do you mean? SandyGeorgia ( Talk) 11:59, 6 January 2020 (UTC)
A meta-analysis of trials of chlortalidone for high blood pressure found that lower doses of chlortalidone (e.g., 12.5 mg daily in ALLHAT study) had maximal blood pressure lowering effect and that higher doses did not lower it more.SandyGeorgia ( Talk) 12:16, 6 January 2020 (UTC)
Part 4: Barkeep49 said:
In the time that I've been working on this conflict I've worked hard to treat all editors with respect. ... But I also remember acutely what it's like when you don't have the sysop flag. I wrote, in a line I had to remove from my ACE statement due to space constraints, "I remember what it’s like to feel put down not or otherwise dismissed because" I wasn't a sysop. We lose so many good editors for so many reasons and I'm sorry that the conditions here are such that we're going to (potentially) lose you. The places you've chosen to contribute in this dispute have been made better because of your contributions. I can only hope you decide Wikipedia remains worth it.
First, thank you Barkeep for the kindness in your response. As to whether Wikipedia remains worth it, I am first and foremost a medical editor, and next, was highly involved in the featured article process. Obviously, I would prefer to contribute medical FAs to Wikipedia, and I can better spend my time IRL if I can't do what I do best here. That's why I'm here, trying to resolve this conflict.
I do believe you have treated everyone equally, and with respect. I'm glad you raised the memory of what it feels like to be treated lesser when you don't have a sysop flag. In my case, I never wanted it, not only because it would be a distraction from contributing content, but because of one of my earliest experiences on Wikipedia. I was attacked by someone claiming to represent "we admins"; [7] the position expressed there was completely fictitious, against every behavioral policy, and I had no recourse but to sit on it. A few months later, Raul appointed me FAC delegate, and a year later, that admin was desysopped. I understood then that a fact of Wikipedia's dispute resolution processes was that it takes a long arbcase and a lot of different instances of misuse of the tools to deal with admin abuse, so we regular editors had best simply accept and live with that reality.
In this instance, although you have treated everyone equally, and "adminning" the MEDLEAD RFC was not your remit (and no other admin came to your assistance), let's look at how non-sysops vs sysops have fared.
Yes, you have treated everyone equally and respectfully. But nonetheless, not everyone ends up being treated equally. I have a tough skin because of years of corralling cats at FAC, and having to stay above it and stay neutral in disputes no matter what was thrown at me. But the effect piles up, and then it is best to take a break for tea, and I'm sorry that occurred on your watch, which has been fair.
The reason it occurred when it did is because I so abhor the effects of backchanneling (something that was well understood during my tenure at FAC, and everyone knew if they cooked up support off-Wiki, I'd shut down their FAC ... and don't even think of emailing me about your FAC unless it is something like a serious COI, which did happen once, and I wrote the arbs). My issue is that if you had had the same conversation with those admins in public view, others would have had the opportunity to point out the items in Point 3, without us coming to this level of awkwardness after the fact. All of this thread could have been avoided; transparency works. I still feel awkward about putting this on this page, but understand why you wanted it here.
I believe Part 5 is now completely summarized back on your talk, but having lost a bunch of this post, I may need to go back and re-read everything to see what I have not addressed.
I think we have a way forward, in addition to the RFC, that doesn't require arb intervention at this point. I believe it has now been revealed that the premise that the ANI close was founded upon was faulty, and we can revisit. That does not mean we should not launch the RFC that we have worked so hard on. Best regards, SandyGeorgia ( Talk) 22:57, 4 January 2020 (UTC)
"We have reams of evidence above that all of the drug price information we have presented to our readers in these 500+ examples is bogus...[removal of conduct related discussion addressed on my talk page]... I have been asking for a month, and we have not yet seen one single example of these database sources being used correctly."This is where I think we've been ill served that the people who feel that pricing information has been used correctly are not participating as actively as those who oppose it. Though I will note that at times both WAID and Trypto have presented modified versions that they feel is policy compliant. Those examples have not been satisfactory to you which is fine. Different Wikipedians can have different interpretations of our policies and guidelines. So I would suggest in fairness that your second sentence is really "we have not yet seen one single example of these database sources being used correctly in my expert opinion" (with the words in bold being my addition). This kind of situation is what the essay WP:POLSILENCE is talking about because everyone reading this should know that the silence here doesn't mean that everyone agrees. And at the end of all of this, that's why we're having an RfC to find out if people agree with you (and the interpretation of policy and guidelines that you and others who share your beliefs cite).That RfC, hopefully, gets us a consensus that lets us start to adjust articles around drug pricing again. In the interim when asking to interpret the embargo I have been attempting to balance "What will keep this conflict from escalating?" with "What will serve our readers?". So adding disputed tags serves our readers but did not, in my evaluation, do enough to serve our readers to outweigh the potential that had to escalate the conflict. Similarly adding to a draft (which obviously won't be seen by most readers) does not offer enough benefit to our readers to outweigh the potential to escalate the conflict. In this case I made the judgement that it could potentially help the conflict (by possibly showing a formulation that some editors would find appealing) while also helping our readers. Of course it could also exacerbate the conflict. That's why I decided this was the time I needed others thinking. Through a mistake filled route we got here (which I've previously addressed and accepted responsibility for) we do have a second uninvolved sysop ( L235) offering their related reasoning that reached the same conclusion. I have, for reasons I explained in more depth in the conduct response on my user talk page, not felt that any history was appropriate to consider in light of the ANI's consensus that there would be
"no rehashing of grievances.". What I read you writing here is that you weigh the competing priorities of service to readers and conflict escalation differently. Which I understand and have and will continue to consider but also does not at this point change my interpretation of that balance. Best, Barkeep49 ( talk) 04:07, 5 January 2020 (UTC)
I have been asking for a month, and we have not yet seen one single example of these database sources being used correctly.That doesn't mean there may not be one; yes, we are hamstrung because of silence of those supporting this data source. Re WP:POLSILENCE, I am not convinced that essay applies here, and I wouldn't even use it at the MEDLEAD RFC, where it might apply-- it's an essay. More significantly and to matters at hand, if we can get the chlorthalidone example moved to its own section, we can continue to explore now one example. SandyGeorgia ( Talk) 17:55, 5 January 2020 (UTC)
Collapsing not that important comment as suggested by Barkeep49.
Nil Einne (
talk)
14:17, 6 January 2020 (UTC)
|
---|
|
Draft RFC at: Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices
Thanks for getting this moving, WAID. I am unclear if we are actually restricted to generics? Why do our questions not apply to all drugs? SandyGeorgia ( Talk) 03:08, 2 January 2020 (UTC)
Colin might want to switch out the sample drugs to include one where it's not even clear what the dosage is or what is being treated. SandyGeorgia ( Talk) 03:10, 2 January 2020 (UTC)
WhatamIdoing, I think there is an initial problem with the "what the source says". We need to back up a bit to see what the source says about the drug, and not just one variant tablet size or formulation: Ethosuximide, Carbamazepine and Mebendazole. We can see from that:
So we need to be up-front that the source offered options, and the editor chose one of them. We need to be careful the examples don't offer illusory easy answers to picking one of them (such as, there's one variant that has lots of suppliers and the others have none, or that the 100mg and 200mg tablets work out the same price by dose anyway). The existing method for choosing a variant isn't foolproof: the diazepam article picked the wrong tablet with no suppliers whereas a different tablet size has lots of suppliers. -- Colin° Talk 11:42, 2 January 2020 (UTC)
[snip]
We are giving the prices and DDD but not actually helping readers with the maths. This matters because sometimes the DDD is being used and sometimes not. With Ethosuximide, if we multiply the 0.1845 250mg by 5 (1250mg DDD) and by 30 we get £27.67 (not sure where 27.77 comes from). With Carbamazepine, we are multiplying these 200mg tablet prices by 5 to get 1000mg and using that for the daily dose. With Mebendazole, we are actually just giving the price of the 100mg tablet, not the DDD of 200mg. So what do we even mean by "dose"? Since the DDD isn't being used here, it may be worth me explaining to you guys using the BNF that you can't read in the USA, which likely focuses on the kinds of parasites we get in the UK
So a "dose" could be the 100mg one-off dose, or the 200mg daily dose for three days, or the 500mg dose you take once. How on earth do we cover this? For many youngsters on Wiki, they may only be familiar with ibuprofen tablets and the contraceptive pill, and unaware that medicine dosage and indications for drugs are complex. Maybe we need a little side-box for each drug, that explains things that Wikipedia generally is forbidden to cover like how the dose is recommended for each indication/patient-group. -- Colin° Talk 11:58, 2 January 2020 (UTC)
Wrt wording "One organization said that they sold ... in 2014" should really be "One organization said that they sell... in 2014". We have price data but no evidence they actually sold any. It might be simpler to call these "organizations" "suppliers", especially as that's what the source calls them, and we will end up discussing buyers and suppliers. -- Colin° Talk 11:42, 2 January 2020 (UTC)
In the Carbamazepine example, you have focused on the suppliers. In fact, the highest price in the article text ($0.24 per day) is taken from one Buyer (SICA: System of Central American Integration). So, to discuss what's gone on in that article text, we need to talk about Suppliers and Buyers, and should really enlighten readers that for example WHO encourage we take the median supplier price, and only consider this representative of an international price if there are many suppliers. We can source this and if necessary quote verbatim. --
Colin°
Talk
11:42, 2 January 2020 (UTC)
In the Mebendazole example, the highest price in the article text ($0.04 per dose), comes from one Buyer (South Africa Department of Health) and in a package of 6 pills, not 1000. -- Colin° Talk 11:42, 2 January 2020 (UTC)
The bits at Additional information can be found at... and Previous discussions on this subject include... still need to be finished. Anyone's welcome to add whatever they want there (or post it here, and I'll add it). I'd really appreciate some help with finding all the relevant things. WhatamIdoing ( talk) 22:27, 2 January 2020 (UTC)
I don't want to be here, but here I am. Please feel free to ignore me, but. I still think there is no need to present three examples from the same database. We will get a result about one database. We could present three examples from the three different sources that have been incorrectly used, and get a broader result. It is the same principle; we have no good data on drug prices anywhere. SandyGeorgia ( Talk) 12:30, 4 January 2020 (UTC)
After a couple days away, I just looked at the RFC page, and it looks ready for launch to me. What work is remaining? Nice work, WAID. SandyGeorgia ( Talk) 18:06, 4 January 2020 (UTC)
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Actually, I'm feeling ignored, too. I think it will be a fiasco if that version of the RfC gets launched. And I've been saying so quite clearly for a long time now. If the consensus really is to do it that way, I won't stand in the way but I also won't refrain from expressing my concerns. I feel like a very small number of editors are, in effect, trying to run out the clock, and aren't really engaging with my expressed concerns, even though I've been revising my suggested version of the format in response to their concerns.
So here is what I would like. I hope that editors who have been watching here, or who have just started watching here, will provide additional opinions about the two draft versions: Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices and User:Tryptofish/Drug prices RfC draft 2.
I'm not asking for editors who have already expressed an opinion to repeat themselves here. I'm hoping that more editors will express their opinions, giving a wider range of input. Thanks. -- Tryptofish ( talk) 20:52, 4 January 2020 (UTC) A part of it struck. -- Tryptofish ( talk) 23:04, 4 January 2020 (UTC)
User:Barkeep49, others, I think things have gone a bit off the track here. This top-level section heading was created to discuss WAID's draft and is now discussing the genesis of and participation at Tryptofish's draft. The former is imo nearly ready to roll and the latter little more than a sketch of an idea (sorry, but true). After almost a month since the ANI was closed and we were asked to form an RFC, it seems we are being dragged back to square one, still scrabbling around to find even one example to put into this second rival draft RFC. It seems we have two styles of RFC:
The point of having the RFC is that we do not claim to know community consensus on this matter, and since polling is not a substitute for discussion, it seems wrong at this point to go for a simple poll, and to offer a false dichotomy that voters must pick positions on. Polling is known to polarise debate and separate voters into adversarial factions, so this would not seem to be a wise option when we have already seen incendiary claims of collusion with Big Pharma to censor Wikipedia and conceal prices from patients. We don't need closed-minds formulating some sound-bite that will destroy the opposition's argument. We need open minds to look honestly and carefully at this complex issue. So complex, that I think we do need to concentrate for now on one source and one region (developing world). I keep saying, please lets choose the least-conflict option, and an RFC that divides Wikipedians into two camps to try to outvote each other is not that, imvho. -- Colin° Talk 21:43, 5 January 2020 (UTC)
Just to make it clear, there was not an "ANI consensus [] that [Colin] should not be the person to formulate the RfC".This is correct. There was one comment in the ANI:
There may very well be problems, but I'm unconvinced that Colin is the best person to be the primary one drafting an RfC to deal with them. Nil Einne; I believe that Colin has acted correctly in respecting the spirit of that one comment, and leaving the actual drafting to other parties. I have full confidence in WAID because of her RFC experience and long-evidenced neutrality in all matters WPMED. She has done a fine job in laying out one part of the problem that needs to be examined before we can move to the broader issue of what to include in MEDMOS, while not giving respondents too much to deal with in one shot. Trypto, I hope you will back off on the term "laughable" and that we can move forward step by step, which is something we've discussed many times on this page. SandyGeorgia ( Talk) 11:15, 6 January 2020 (UTC)
Let me be clear: I am not going to write any RFC, and I don't think James should write one either.: [13] And then you told me just above that
the ANI indicated Colin shouldn't do the drafting, which I see now that you have struck: [14]. Once you told me that, I figured that I just hadn't remembered it. That's what it came from. -- Tryptofish ( talk) 20:49, 6 January 2020 (UTC)
;-)
The narrower RFC that I have drafted at
Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices was not originally my idea, but I no longer remember who first mentioned it.
|
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Barkeep49 I know you said that certain concerns should be done on a user talk page rather than a guideline talk page, but I'm prevented from doing that. I have to say I'm finding these allegations by Tryptofish to be rather against our assume good faith policy, and not conducive to collaborative working. Trypto, here you are frustrated that I haven't helped fill in your skeleton RFC and suggest I'm deliberately leaving it bare in order to comment that it is unfinished. The allegation that folk are "rigging the RFC to get a desired result", and query to WAID about her response to a potential complaint raised "that the RfC is hopelessly biased and should be discarded" do not make me feel comfortable. Your "let's not let the facts get in the way" is fightin' talk, Trypto, please tone it down. Some of us have been at this since October! Let me be clear: I am not going to write any RFC, and I don't think James should write one either. I can express my opinions about the content/focus of the draft RFCs and you guys can accept or reject it same as with any other editor here. I think we are all rather tired, and frustrated at the lack of participation by experienced editors in helping drafting any RFC, but lets please collaborate and compromise, not fight among ourselves. -- Colin° Talk 22:45, 4 January 2020 (UTC)
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Barkeep's Background
|
---|
A few days ago, Colin quite reasonably asked me for some examples of good RfCs to use as a model. I spent some time today looking into past MOS (and related) type RfCs and also asked a few people with experience closing Wikipedia related RfCs for examples. I am reluctant to share any because I don't know that they will actually help move the conversation forward. What seems to happen is that someone puts forth an idea and it gets criticized for either structure or wording. Much/all of the criticism is fair (and I've done it) but ultimately it means we aren't getting any closer to an RfC. Above Colin expressed frustration with the status quo of the information being included in hundreds of articles at the moment. A few others seem to be feeling this as well. Others are no doubt being frustrated at the inability to add the information in new places. The way past this is to get this RfC completed. Editors of all stripes need to have some faith in our process. The RfC question itself is not the right place to put forward compelling arguments about why pricing is/isn't appropriate. Instead that can happen during the RfC or in some sort of supporting material. If the goal is for the RfC is to solve every pricing related question the whole RfC is going to fail and nothing will be decided. Instead the decision should be made, by the people here who care most, about what's most important. To everyone I remind it is possible that not everything that's important will get decided by this RfC. So what's most important to decide? Some willingness to to accept that important issues won't be decided right away needs to be tolerated in order for there to be any chance of moving this forward and it seems clear that this needs to move forward. If a person's answer to "what weaknesses/compromises can I live with?" is nothing then that person is not going to be able to successfully participate in formulating this RfC.The good news is that no one seems to yet be at that point. The further good news is that multiple people are seeing progress being made. In rereading this talk page it seems like there is some level of agreement behind a single question (that can be answered with a support or an oppose). People seem OK with coming back to the details later. This is helpful because a single question also seems most likely to achieve consensus from the community. If we can't get consensus around a single question I think it could also give us insight into what the multiple questions have to be at this point and we can, if necessary (though I think it might not be) examine different formats for multiple questions. As such I am going to propose the following. |
I am suggesting we focus, for now, on trying to perfect a single question about pricing (or whatever your preferred term is). I am also going to ask that people to follow my lead and collapse extended content, leaving only the most important information visible. Say as much as you want, but let's make it easy for people to navigate. Thoughts? Barkeep49 ( talk) 21:22, 23 December 2019 (UTC)
So far the following single questions have been proposed:
Do you think that a pharmaceutical drug has one price that can be expressed in dollars and cents, for each region such as US, UK, developing world, and this is information that Wikipedia drug articles should include and can routinely be sourced by editors without original research.
— User:Colin
Should Wikipedia articles contain information about the cost of medications?
— User:Doc James
Please take one (or both) of these and wordsmith them to your heart's content. And if you want to explain, at length, why your version is good, or issues you see with someone else's proposal, feel free but again please consider collapsing those comments. Barkeep49 ( talk) 21:22, 23 December 2019 (UTC)
SG's Background
|
---|
|
Do you think that any individual pharmaceutical drug has one price that can be expressed in a given currency, for any region such as the US, UK, or the developing world?
— User:SandyGeorgia
Followup: What kinds of sources for pharmaceutical drug prices can be cited by editors, without original research, for Wikipedia drug articles?
— User:SandyGeorgia
Thanks for all your hard work, Barkeep49; this amount of effort is beyond the call of duty. SandyGeorgia ( Talk) 01:43, 24 December 2019 (UTC)
— kashmīrī TALK 02:22, 24 December 2019 (UTC)For articles discussing medications, do you think that Wikipedia can, reliably and without original research, source medication prices as used in various regions of the world; and if found, should convert them to a common currency and include in articles?
we don't want prices in South African rands, RTGS dollars or Chinese yuans, do we?- Why not? Wikipedia is an internationally targeted encyclopedia, what currencies do we want prices in?
Inconsequential formatting change
|
---|
|
I think that the main question to be settled is how much WEIGHT do we (by default) put on sources about prices (as understood in the dollars-and-cents model, not the general how-many-people-can-afford-that sense). Does it fall into (or near) the category of basic information that User:Bluerasberry calls Wikipedia:Defining data, in which case we need to include something about it whenever reliable sources permit us to say anything at all? Or, alternatively, is this content something that we should normally not include, and only mention when we have especially good sources (e.g., multiple high-quality sources that discuss the price at length). Here are some examples that we might consider:
Subject | Source type | If we put a lot of weight on prices | If we put less weight on prices | Notes |
---|---|---|---|---|
Insulin | Many long articles in news media and academic journals, including claims of price gouging and people dying because they couldn't afford the drug | In the US, between 1987 and 2017, the wholesale price of long-acting insulin increased from US $170 to $1,400 per vial.[1] | In the US, between 1987 and 2017, the wholesale price of long-acting insulin increased from US $170 to $1,400 per vial.[1] | |
Valproate | A 2017 peer-reviewed journal article, and some data points in various databases (i.e., independent primary sources) | According to estimates published in The BMJ in 2017 for drugs on the WHO Model List of Essential Medicines, the cost of manufacturing the active ingredient in this drug in India, is approximately a couple of US cents per pill.[2] | (Nothing – this is a single primary source) | The cost of manufacturing the active ingredient is reasonably consistent worldwide. India is the biggest producer of these generic small-molecule drugs. But no retailer or consumer buys just the active ingredient. |
Denosumab | An article in a pharmacy industry magazine (independent and possibly secondary) | Shortly after its original approval in 2010, Medicaid's US average wholesale price for a 60-mg prefilled syringe of denosumab was reported at US$990 per dose, with two doses expected each year to treat osteoporosis.[3] | (Nothing here, but maybe something in the manufacturer's article) | Just one dose (of two for this drug), in just one country, at just one point in time, using just one metric (of many). |
Golodirsen | An article in a biotech business magazine reporting on an Earnings call (independent and primary) | In the days after Sarepta Therapeutics received permission from the US FDA to market the drug, the net annual cost was estimated to run around US$300,000 per treated patient, assuming the patient was a child weighing 25 kg (55 pounds).[4] | (Nothing here, but maybe something in the manufacturer's article) | No actual sales and little non-business coverage at that point, but high-cost drugs tend to attract attention, so maybe more sources would appear later, at which point it might be treated more like the Insulin example. |
Abacavir | A routine entry in a government database | According to the Drug Pricing and Payment database maintained by the US Centers for Medicare and Medicaid Services, the National Average Drug Acquisition Cost for 300 mg pills of abacavir was US$0.77 each in December 2019 in the US. | (Nothing, because it's a primary source) |
All of these examples have been mentioned in the discussions leading up to this point. If anyone feels like any of the examples are misrepresented, please let me know. WhatamIdoing ( talk) 05:41, 24 December 2019 (UTC)
Colin's Background
|
---|
I think we have rather forgotten why we are having this RFC.
I hope a pattern is emerging. We aren't having an RFC about some new or alternative idea for prices in Wikipedia articles. We are having an RFC about the actual current prices in actual hundreds of articles. We're having an RFC because of an impasse between two editors. And we're having an RFC because WP:MED has completely avoided making direct explicit criticisms of that text or of fixing any problems in the past three years. WP:MED is clearly not going to fix this and we need input from the wider community and neutral editors to contribute. We need to offer a question that directly resolves this matter, rather than creating new ones. -- Colin° Talk 10:01, 24 December 2019 (UTC)
|
I think question A (by Colin) is a necessary and sufficient condition for the current text in hundreds of drug articles to be kept. Splitting in two like Sandy proposes weakens this, especially the open question B which may not likely lead to anything other than a random mix of opinions. The problem with a "What kinds of sources" question is that it always depends what you use if for and "for pharmaceutical drug prices in drug articles" is not specific enough. Many people like to view sources as adjectives. So "MSH" or "data.medicaid.gov" or "WHO" or "BNF" will be viewed as "reliable sources" and as "secondary sources" (they aren't the primary source of their data). So I suspect we'd just get comments like "Must use high quality secondary sources like the BNF" without stopping to think that the BNF may list 30 prices from 10 manufacturers for a drug, or even working out what the two prices the BNF list for each record actually mean. I'll try to find some representative article texts later. Essentially we want an "Are you happy with this?" question around existing practice. -- Colin° Talk 10:21, 24 December 2019 (UTC)
Putting this here at the bottom, to encompass much and good feedback above. Please, people, come on ... take Barkeep's suggestion and put up concrete suggestions so we can start wordsmithing and discussing specifics. Once the proposals are up, we can see the issues and refine. I am at the limit of my wordsmithing ability, and despair has set in; length has again taken over this discussion, and we have nothing concrete. Trypto and Peter and Ronz, give it a go even if you aren't yet fully satisfied with what you might intially propose. SandyGeorgia ( Talk) 19:02, 25 December 2019 (UTC)
An article should not include product pricing or availability information unless there is an independent source and a justified reason for the mention. Encyclopedic significance may be indicated if mainstream media sources (not just product reviews) provide commentary on these details instead of just passing mention. Prices and product availability can vary widely from place to place and over time. Wikipedia is not a price comparison service to compare the prices of competing products, or the prices and availability of a single product from different vendors or retailers
"And I agree with that. All the sources being used to discuss prices are independent. And we have sources that justify the importance of pricing information". We have some editors who think prices concerns in national newspapers is what counts as independent discussion and WP:WEIGHT to include. Others claim that multiple drug databases listing prices satisfies WP:WEIGHT and finding some obscure internal memo on the price of X is sufficient discussion, or that general concern about drug pricing is enough to justify the inclusion in every single drug article. So we need a question that cuts through all that to make a clear consensus. If we just end up with a reworded WP:NOTPRICES that is immediately ignored/evaded we haven't achieved anything.
Should reliably sourced and unambiguous dosage prices be routinely presented in articles on drugs (medications) in a prominent position (lead or infobox)
— User:Pbsouthwood
"The wholesale cost in the developing world is about US$0.01 per dose as of 2014.". Again there are a whole host of problems with this because the source lists only the Buyer price in the Dominican Republic and in Peru, and lists no suppliers at all. This should have run alarm bells for such a huge drug. In fact the 10mg tablet would appear to be uncommonly used (hence no suppliers at all, and in 2015 only Peru was a Buyer) and the 5mg tablet here with eight suppliers is far more reasonable. But what is a "dose". Would someone writing "costs ___ for a 5mg dose" and citing this be "reliably sourced and unambiguous" in your book? The source does not indicate which tablet size to pick or what is a "dose". Nor, for other drugs, does it suggest whether to use enteric coated tablets, or suspensions. The MSH is a "reliable source" for some things (nobody is doubting it is generally likely to be correct about the prices it lists) but not a "good source" for others (the Buyer prices in Peru and DR are not considered representative of the "developing world" according to ANY recommended usage of MSH). Other articles give a cost per day or per month or per treatment. Are they "dosage prices"? By "unambiguous" does one need to include the exact indication too. For example, the BNF gives all sorts of dose options for various indications, many in a range. How would we pick which indication? And if we picked one ("Muscle spasm of varied aetiology") the adult dose is
"2–15 mg daily in divided doses, then increased if necessary to 60 mg daily, adjusted according to response, dose only increased in spastic conditions."How does that translate to a "dosage prices" in an article? The BNF prices are here.
Would someone writing "costs ___ for a 5mg dose" and citing this be "reliably sourced and unambiguous" in your book?
I am concerned that our questions must be asked in a way that addresses the NOR aspect of these drug databases, the WEIGHT aspect of whether we should include prices at all, and the LEAD aspect of whether they should be in the lead. If we really have so many core policies being challenged, what the heck. NOR, NOT and WEIGHT are policy; why are WikiProject guidelines and practices challenging policy and why is that not being done with RFCs on the core policies? I am going to end up dissatisfied if we don't have questions that will get us to the core problems. If we still had an RFC/U process for user conduct, we would be asking these questions there. SandyGeorgia ( Talk) 14:56, 26 December 2019 (UTC)
So, let's consider the answers that F will generate, and how the closer will interpret those? F presumes a baseline understanding of and achknowledgement of reliably sourced and unambiguous dosage prices. We don't have that in this price dilemma. If we did, we wouldn't be here. So, what will the closing admin do with the "ILikeIt", "Me, too, per editor-so-and-so" responses that will not engage the core questions and policies? The RFC MEDLEAD shows we will get "because I like it" responses. We need to be highly specific in our questions about the core policies: NOR, WEIGHT, NOT. With the MEDLEAD RFC, a closing admin can argue that guidelines are flexible and can be ignored. In that case, with respondents not engaging the core questions, we end up with protracted local discussions to determine consensus on individual articles, where one group will argue LEAD and another group will argue MEDLEAD. We end up with articles that cannot be taken to FAC, because you can't please two masters. This is really not a big deal, because essentially no one at WPMED is attempting to write complete articles anymore anyway, and no one is maintaining most of the project's Featured articles.
Unlike the guideline LEAD, on core policy questions, the problem cannot be so easily overlooked. In this case, if we end up with an inconclusive RFC where respondents do not engage policy because we haven't asked the questions with great specificity, what's next? Protracted local disputes end up at arbcom. SandyGeorgia ( Talk) 16:22, 26 December 2019 (UTC)
I really think this is important: NOR is not DUE.
Deciding which entry to cite in a database is not a NOR violation. It's (possibly) a DUE violation, but an accurate description of the content published in a reliable source is never NOR. This means that if you look up wonderpam in The Database, and it says "100 mg pill – generic – Specific Price Type – US$0.10 – December 2019 – UK", then writing "According to The Database, the Specific Price Type of a 100 mg pill of generic wonderpam was US$0.10 in December 2019 in the UK" is not original research. That's actually what the published reliable source said; therefore, that's not NOR.
Now, while that statement is not a NOR violation, NOTPRICE suggests that it's also probably not something that we want. It's possibly unencyclopedic, and it's very likely UNDUE emphasis (why that one size, that one date, that one country?), even though it's not actually original research. I don't think that we'll get a sensible RFC response if we go to editors and say "He copied this information straight out of a single reliable source – that's a NOR violation, right?" NOR means "material—such as facts, allegations, and ideas—for which no reliable, published sources exist". If you're copying it straight out of a single reliable source, then it's not NOR. IMO we need to stop talking about NOR (which is either not a problem at all, or is a problem that can be fixed by copyediting) and focus this discussion on DUE.
"Focusing on DUE" IMO means that we ask editors how much emphasis we should put on this subject. "How much emphasis" is partly subjective. Yes, you have to have the sources, but if something is "always" DUE, then you can/should write a weak claim from whatever source you can get.
We need to know whether editors want minimal emphasis (in which case, we remove a number of existing statements, or at least move them out of the lead), or whether they want significant emphasis (in which case, MEDMOS can provide information about how to write non-NOR statements), or something in between.
It would be good to educate the respondents about how complex the subject area is, but if they want significant emphasis on this, then the feasibility of implementing their goals in any given case is not the primary factor in writing our advice. After all, we put a huge amount of weight on the dates and locations of people's births, even though we know that can't source birthdates and locations for every single biography. Putting a huge amount of weight on it just means that if you've got any source at all, even if it's just one unimpressive primary source, then you include whatever you've got, with whatever WP:INTEXT attribution and careful description that allows readers to understand the limits of the source.
When this started, I gave a pair of examples in MEDMOS:
If editors want prices at all costs, we could add one that shows how to respect NOR while still including database-derived prices:
We could also add an example from news media:
I know those statements can be sourced for some medications without transgressing NOR. IMO what we need is to know whether editors actually want us to do that.
With that in mind, I think that the question to ask is:
How much WP:WEIGHT should be put on the prices of pharmaceutical drugs?
This question can (and IMO should) be followed by examples (the despair-inducing table) and explanations (the impossibility of finding the One True™ Price for a drug that sells in 190 countries in six different doses and four common formulations under at least 90 brand names) and alternatives (we could skip dollars-and-cents and instead try to source a comment about affordable/expensive), and comparsions to similar subjects (e.g., how NOTPRICE is applied to other products), but I think that this is the most basic question to ask.
I do not think that we should be asking a yes/no question. I think editors should be encouraged to respond with both an overall view related to inclusion (e.g., always include, usually include, usually omit, only include under extraordinary circumstances, only for generic drugs, only for WHO Essential Medicines, only for drugs under patent protection, only for drugs with unusually high prices – whatever editors actually want) and with an idea of how to include (e.g., brief mention in the lead vs a whole paragraph or section, current prices vs original prices, etc.). WhatamIdoing ( talk) 19:14, 27 December 2019 (UTC)
WhatamIdoing, sorry but I think you are totally wrong, because you keep having a strawman argument about fictitious possible price/cost/affordability statements some fictitious article might contain. We aren't having an RFC because of fictitious possible article text. Of the 500+ drug articles that currently display prices, all of them required original research to make the statement they do. When we chose one unnamed tablet to represent "the cost", that was indeed original research and when we multiplied by a "dose" that was also original research. Have you looked at the medicaid source links? They cite a "prices for week xx in 2018" database of tens of thousands of records. You need to then, by hand, filter the results to the drug the article is about and then you need to look at different formulations and tablet sizes and try to reverse engineer which one was picked to get $50.45 a month or whatever we claimed. And of course the medicaid site doesn't mention a dose at all, so no "intelligent reader" could possibly work it out from that source -- the very definition of original research.
And we also made claims that are not supported by the source at all, like "the wholesale cost in the developing world" citing one buyer price in Costa Rica, or claiming the price ranged from $x to $y when that is also not supported by the source. None of actual articles formulate the price statement like you did. If they did, then you could indeed make the argument that it was undue weight to mention one tablet size from one drug manufacturer in one country. But they don't and the difference is not solvable, as you put it, by "copyediting".
We nearly always give a price per dose, per day or month (which also require picking a dose) or per treatment (which requires an indication and dose). But we always don't mention what that dose is nor do we mention what the indication is. Terbinafine was one example above, Aciclovir another I spotted today, where the article does not state whether the costs are for a cream for cold sores or a tablet for shingles or post-transplant infection of cytomegalovirus. Yes there would be a WP:WEIGHT problem with explicitly giving the price for just one indication, and there would be a WEIGHT problem if we were explicit about the formulation/supplier/etc we used for our maths. But we aren't even specifying the indication, never mind the other things, so our problem isn't WEIGHT but just being dishonest with our readers in claiming there is One Price. The "we could be as specific as The Database is" argument is a false one to make and knock down, because nobody is proposing it. As you say yourself, just giving an example of all the permutations should be despair-inducing. We need to also remember MEDMOS prevents us from stating drug dose information in articles (for good reason) so we can't even explain to our readers why we picked the 250mg tablet.
WhatamIdoing, I have complained about the horrendous problems with the lead text in 500+ drug articles, and I think an RFC that appreciates the problems with that text will result in all those lead prices sourced to databases texts being removed from all 500+ drug articles. Please, the RFC must resolve the conflict over existing article text, not generate imaginary conflicts with imaginary texts. I don't think your WEIGHT question resolves this conflict at all. It isn't even in my mind a sensible question to ask. WP:WEIGHT is determined by reading the body of literature on the article topic, not by Wikipedians expressing a personal opinion. -- Colin° Talk 12:54, 28 December 2019 (UTC)
"The wholesale cost in the developing world as of 2014 is between US$0.36 and US$0.83 per day. As of 2016 the wholesale cost for a typical month of medication in the United States is US$70.50". [18] [19] The DDD on the MSH site says 600mg daily dose, as does my BNF and Drugs.com. So that's two 300mg tablets a day as the developing world prices agree. But the US price of $70.50 is approximately the $70.35 I get by multiplying a 300mg tablet price of $2.34487 by 30. The actual 2016 US monthly price should be $140.69. (Why we have one price per day and another per month is beyond my understanding). Leaving aside that neither MSH nor Medicaid state what the typical therapeutic dose is, for us to do original research on, this is just mathematical incompetence. And it is very typical. I am repeatedly seeing prices citing the BNF that assume a pack of 28 tablets is a "month's cost", totally ignoring that a patient might taken more than one tablet a day. So, the evidence does rather suggest that disallowing original research is a jolly good idea, because we are crap at it. And then we see that the price hasn't been updated since 2016. As your citation shows, the equivalent price in 2019 for 60 tablets would be $46.45, which is about $100 a month less. No small change that, but nobody it seems, is interested in either the price in 2016 being right, or giving the right price for 2019. -- Colin° Talk 21:40, 28 December 2019 (UTC)
In terms of presenting this question, I think it needs a bit of explanation. The straight-up question is "How much weight?", but after that, some explanation is necessary. One way (of many ways) might be to explain the context, and then offer some considerations. It could look something like this:
All Wikipedia articles should present information with WP:Due weight. Generally, this means that the more our reliable sources talk about an aspect, the more attention that aspect should get in the Wikipedia article. However, there is some information that is considered so important that it is included whenever possible. For example, in a biography, we include information about the subject's birthdate whenever possible, but we normally mention the subject's hair color only if reliable sources dwell on the person's appearance.
Drug pricing and affordability is a significant area of discussion in reliable sources, but this discussion is almost always held at a general level, and does not extend down to individual products. The prices of individual pharmaceutical products vary so widely by place, time, dose, and other factors that general claims, such as "the price worldwide" or "the price in developing countries", are almost always incorrect. It is, however, frequently possible to source a statement about what a particular metric yields for the price of a particular size of a particular drug from a particular manufacturer in a single country on a given date.
Editors who work on medicine-related articles have recognized that much of the information about drug prices currently in Wikipedia articles is not an example of our best work. Much of it is outdated or otherwise incorrect. We want to fix it, but we have not been able to agree on the best approach yet. On the one hand, the cost of a drug affects whether people can get it at all, so some editors believe we should always include whatever we can source. Other editors believe that pharmaceutical drugs should be treated like any other manufactured product, and that means no prices unless we have multiple reliable sources discussing the price of that particular product in depth (as we do for some, usually because of very high costs). Editors fall across the whole spectrum from maximizing inclusion and prominence, through the middle grounds, to the opposite side of including as little price information as possible. All of us want to know other editors think, so we're asking you: How much weight should we put on drug prices?
To explain some of the positions, a few editors have offered background information that may be useful to you. We hope that you will join us in a conversation about the best way to handle this subject area.
(Collapsed – It's incredibly important) (Collapsed – You wouldn't believe how complicated and useless this is) (Collapsed – The middle road is WHO/HAI affordability, not prices) (Collapsed – People should care about pricing, not prices) (Collapsed – Where and how we mention prices matter more than whether we do) (Collapsed – What we could actually source is unencyclopedic) (Collapsed – Whatever other ideas/positions/recommendations I've forgotten)
My suggested "collapsed" sections could be written by different people, in the hope that editors would read more than just the headlines. Yes, it's long. That's not necessarily a showstopper. The important question is, if we asked this, do you think that we would get responses that would help us figure out how to clean up these articles? WhatamIdoing ( talk) 06:53, 29 December 2019 (UTC)
SG's attempt at a new start over, abandon hope all ye who enter here
|
---|
IMO, the reason we are unable to formulate an RFC question or questions is that the task we are undertaking is the wrong one: we are attempting to formulate a general RFC to address what is in fact a very specific dilemma relating to very specific databases, when the answers to our sourcing and content dilemma are already addressed by policy. We have no other example anywhere, after weeks of discussion, of any other instances of drug pricing in articles presenting a problem. There has been no problem except the database-style sourcing. We are attempting to generate questions that will get respondents to read and respond to what is (should be) a policy question, but we will get "because I like it" responses that will give us nothing useful as result. That is why we are here. Everyone who has participated in this discussion knows how to add price data according to WP:V, WP:NOR, WP:WEIGHT, WP:NOT and WP:LEAD. Why don't we forget all the general questions we are trying to ask to solve a non-existent general problem, and instead just get straight to the specific problem? Put up one example of database-sourced text (I have repeatedly asked the database advocates to give us the strongest example, and none has been produced) and simply ask if this text is supported according to policy, V, NOR, NOT, WEIGHT, LEAD. Then each respondent will lay out arguments of why it does or does not breach each policy. We are spinning our wheels trying to solve a non-existent problem, as if this has been a generalized problem across all drug articles. We have one problem only; over 500 articles using a database to source text. SandyGeorgia ( Talk) 12:57, 29 December 2019 (UTC) |
Do these examples of pharmaceutical drug prices comply with Wikipedia policies on verifiability, due weight, no original research, and what Wikipedia is not? If so, should this text be inserted into leads of articles?
But the examples need to be varied to included the other drug databases, and the other kinds of problems presented; I pulled these samples from one section above, but the three of them were to demonstrate one issue. The other kinds of examples should be give in place of two of these. SandyGeorgia ( Talk) 14:23, 29 December 2019 (UTC)
I had wondered a similar thing to Sandy's idea: that we take each of the four(?) price-database sources and look at how they are being used and then together reach an agreement about the problems with them, what could be said using them, and what shouldn't be said using them. I actually remain optimistic that much of that analysis/discussion/conclusion could simply be done if we get a good-faith article-experienced bunch of editors to simply work together. All the previous discussion on specific problems felt like it was only me and James and that didn't work for various reasons. I'm not rejected the idea of an RFC, but it would be great to clear away a lot of the crap first, and there really is an awful lot of pretty straightforward crap we could eliminate IMO quite quickly, and focus an RFC on asking the community about price statements that are actually source->text honest and policy-legal. I would be much more confident that such an RFC would be successful in its goals (and to be honest, presenting all the awful prices to the entire community right now would IMO seriously dent WP:MED's credibility).
A mix of family priorities, the latest Star Wars film, and going back to work mean I've not yet finished some of stuff I was working on that demonstrate current text and current problems. I do hope I get a chance to put them up very soon. I think then it will be good if we can all see an honest full selection of drug prices in articles (rather than anyone accuse of picking a hard/easy example) and also quite a number of drugs where the price is genuinely notable and editors could make a really good effort to polish some excellent guideline text on how to present that to readers.
Barkeep49, could we have some kind of moderated workshop to focus on e.g. one source at a time. Advertised to the community. We want participation from wise owls or diligent investigators, rather than just ask for a big mob to vote all at once. I don't think we require medical experts at all, so please nobody rule themselves out on that regard. This is really straightforward source->text analysis. The only kind of ability I can see being useful is an appreciation of statistics to the degree that one can't make general statements from few data-points and to identify the weaknesses in one's data. But that is elementary stuff. We could do this workshop on this page even, and do one source at a time. I would certainly like all the current participants to help, but there is one obvious name who is not currently engaging in discussions, and who's participation is essential. -- Colin° Talk 10:06, 30 December 2019 (UTC)
These examples of pharmaceutical drug prices have been taken from the leads of some articles. Do they comply with Wikipedia policies on verifiability, due weight, no original research, and what Wikipedia is not?
I would like a bit more time to prepare some data pages like I've linked below. There's also a couple of wiki articles on price-related topics that still need a bit of work. I'm keen that we present real data rather than appear to cherry pick. As Sandy notes below, for many of the sources, other editors will not easily grasp how on earth the price was derived from the source. It has taken a while for us to figure this out (while picking jaws up from the floor at the amount of original research and arbitrary choices made). WhatamIdoing you ask "Do they comply...."? Is there anything about this that the group here currently discussing prices disagree on or don't know the answer to? I still think a workshop could be a better approach to tackle and resolve the basic stuff that isn't opinions about what is or isn't encyclopaedic or what does or does not belong in a lead. Those are questions we could ask an RFC, but there are basic mistakes with all the texts & sources that really we don't need to ask the community. Do we? Having an RFC to resolve issues with what one editor alone has written and defends is a bit weird if you ask me. If you ask the above question, it is admitting that WP:MED has no competence to discuss, reach consensus and write honest source-based facts about drug prices. That it hasn't a clue whether the prices in the articles are good or bad and needs some help from Pokemon editors and Historical Fiction writers. And while I think that has been true, that WP:MED has been incompetent here and has failed to address or even examine the problem over many years, it doesn't need to be. -- Colin° Talk 14:24, 31 December 2019 (UTC)
"Having an RFC to resolve issues with what one editor alone has written and defends is a bit weird if you ask me."which takes a jab at an editor (who while unnamed is clear to us who've been following this) and which could have been omitted without diminishing the larger point. The broader community decided that the way forward was through the RfC process, a way of gaining binding consensus. Any consensus reached without an RfC will need to include the consent of editors not currently participating in the discussion but who are invested in the outcome. Unless those editors choose, voluntarily because this is Wikipedia and we all have options afforded to us by being volunteers, to agree to that consensus then it will need to go to RfC. I wouldn't say that the chances are 0 of finding consensus without an RfC but they are slim. And if we can't get to an RfC then the only option is to focus on the behavior issues first (through ArbCom) in hopes that this then creates an atmosphere where consensus can be found on the content question. And even then we still might end up with an RfC. Best, Barkeep49 ( talk) 17:43, 31 December 2019 (UTC)
I've posted a reminder of this discussion at WT:MED#Plans for RfC about drug pricing. -- Tryptofish ( talk) 22:12, 3 January 2020 (UTC)
I have created User:Colin/ExistingPrices that is an automated extract of drug prices from drug articles. I got the list of drug articles by looking for external links to the MSH price guide, the Drugs.com price pages, the Medicaid NADAC pages, or referred to the BNF. It isn't all the drugs, but it 530 is good sized sample. I then extracted lines containing the word "price" or "cost" and did a bit of hand-editing on the result.-- Colin° Talk 17:24, 30 December 2019 (UTC)
<references />
tag at the end of each ===Example===? I'm pretty sure that a simple regex find-and-replace across the page would do it, but I wasn't able to figure out the right combination. The main ref tag was updated to auto-limit itself to only the stuff in between the current one and the previous one, so that will get us the refs in each section (so people are more likely to look at them) without exceeding the template limits.
WhatamIdoing (
talk)
20:01, 30 December 2019 (UTC)
I have also created User:Colin/MSHData which lists every MSH Price Guide reference along with the data year cited and the number of suppliers and buyers. In the WHO/HAI price survey methodology, how representative reference prices are depends on the number of suppliers quoting for each product. Because of this, they focus on a small set (14 or 18 products) that have good supplier data. We can see that 30% of our drug citations have no suppliers at all, yet we claim a price in "the developing world". A further 28% only have one or two suppliers, which makes the claim to be representative of "the developing world" a tenuous one. The majority (58%) of our MSH citations for "the developing world" refer to fewer than three suppliers. While some suppliers are international in scope, many target a single country or even just one part of a country. Nearly all (92%) of our MSH prices are from 2014, five years ago. The remainder are from 2015. The guide used to be updated annually but has not been updated since 2015. -- Colin° Talk 23:14, 30 December 2019 (UTC)
Sandy the prices from the Tarascon Pocket Pharmacopoeia are based on a $, $$, $$$, $$$, $$$$$ pricing symbol much like your holiday guidebook indicates if a restaurant is a cheap-eat or an expensive night out. I complained about it earlier at WT:MED. Here's what they the book says about its symbols: (the underline italics is theirs)
So whenever you see "is inexpensive" or "under $25" that came from a "$". If you see "between $25 and $50" that came from "$$". And so on. If you see "more than $200" that came from $$$$$. So even if the drug costs an eye-popping $9000 a dose, we'll just say "more than $200". And as you point out, the majority of drugs are "under $25" even if actually they are just a few dollars. None of the values 25, 50, 100, 150 and 200 appear in the source-data for the wholesale price of those drugs. Those values are all artefacts of Tarascon's price grouping into $ symbols. Our readers couldn't give a damn about Tarascon's price grouping. Reverse-engineering a $ into "less than $25" is a heinous crime. Saying a thousand-dollar drug is "more than $200" is a heinous crime. -- Colin° Talk 13:58, 31 December 2019 (UTC)
In thinking about how this relates to or affects the general question we've posed, I decided to look at the other med we deal with most commonly in the clinic where I am an interpreter. I was not as familiar with metformin pricing, because we are often giving away free samples. (Actually, many patients have their family send metformin from Mexico.) So, I decided to check that one (diabetes). Wikipedia has:
What remains astounding about this formulaic editing is that, in one demonstrable case where we SHOULD have information in the lead about price ( epipen), there is NONE. We could give that as an example, but we can't edit prices right now. (If we decide we need that as a good counter-example, we can do a mock-up.)
In re-thinking how all of this impacts the formulation of the RFC questions, I am coming back to the lead problem, which must be dealt with.
These examples of pharmaceutical drug prices have been taken from the leads of some articles. Do they comply with Wikipedia policies on verifiability, due weight, no original research, and what Wikipedia is not, and do they reflect the guideline on leads?
What else needs to be done before the RFC is officially launched (other than waiting a couple more days)? WhatamIdoing ( talk) 00:03, 1 January 2020 (UTC)
Here are the ones I put up earlier ... we have evolved :)
SandyGeorgia ( Talk) 00:53, 1 January 2020 (UTC)
I had written a long update to the community spurred in equal measures by how close we are to the tipping point here (either towards success or towards a failure to formulate an RfC) and QuackGuru expressing a desire to appeal an aspect of the prohibition. Quack has now said they're not going to appeal so with half the justification for the update gone I undid it.
Barkeep's Summary at AN
|
---|
A little over three weeks ago I closed a long and unsurprisingly acrimonious ANI thread relating to the behavior of several editors. The ANI discussion also had heavy elements of a content dispute around what should be or not be included in the Medicine Manual of Style page. The major finding was that an RfC was needed. In the time since I (as penance for closing that discussion) have been helping interested editors move towards an RfC. I am updating the community now both because I think we're going to be at a tipping point soon as to whether an RfC will ever get crafted (I'm hopeful but it's hardly a sure thing) and because an editor has expressed desire to appeal to the community for an exemption to one area of that close which I expect will be forthcoming soon. All are of course welcome to read the whole long discussion but here's my summary of major points since then:
While all of the editors participating are incredibly skilled and knowledgeable about the topic – far more so than I – sometimes that understanding of complexity sprawls the discussion in interesting, relevant, and important to the topic ways but not necessarily ways that are helping lead to the RfC. I am hopeful that this update is helpful when considering the appeal that is about to come and in the interests of having some more uninvolved editors who can help move the RfC to launch. Best, Barkeep49 ( talk) 01:08, 1 January 2020 (UTC) |
If this is going to get to RfC the more focus we can have on the specific examples to be used the better. Naming is not unimportant but if we can't finish finding the examples for H/I/J the name won't matter. I suspect that this next week is going to be all the difference as to whether we will get the RfC launched and have it find a consensus (still my hope and well with-in our grasp) or not. Best, 01:32, 1 January 2020 (UTC)
"Sure we could start with "do you think the approximate price of a medication can be estimated for various regions of the world?" And "Should Wikipedia articles contain information about the cost of medications?". Furthermore I think it is relevant that when WP:NOTPRICES was quoted earlier, James claimed to agree with it, despite clearly disagreeing with others as to what it means. I note at Talk:Ivermectin#Price that Seraphimblade wrote
"Pricing, per policy, is not permitted in articles, with rare exceptions when the price is a significant part of something's notability or is very extensively commented on (not just mentioned).". Ronz linked here saying
"All discussion so far supports removal"and James yesterday replied
"Ah lots of discussion supports keeping it.". James's two questions aren't a million miles away from my Question A: can we establish "the price" "for various regions of the world" and "should articles contain them" (though most of us want a "while obeying policy" included in that question). Rather than addressing that bigger question, we seem now to be focusing on chipping away at the prices, either by establishing each source shouldn't be used like it has been (and possibly discover if there is any way it can be used), or get community agreement that the current examples fail policy. Maybe that is the correct approach, but we need to be clear that the chipping-away RFCs are not asking the-big-question. Even if we conclude that our uses of sources A, B, C and D are all awful and should never have been permitted and must be swiftly removed, we still haven't resolved WP:NOTPRICES because some editors read it to mean one thing and others read it to mean another. -- Colin° Talk 11:22, 1 January 2020 (UTC)
I think we have to decide next who is going to be the editor to start chunking text in where. My brain is linear. I am to a point where I am stalled until I see something on a page. And we know we don't want me doing the writing. SandyGeorgia ( Talk) 20:21, 1 January 2020 (UTC)
Barkeep used the formulation of good news/bad news, and I'm going to do my own version of it. First of all, I think it's entirely due to say out loud that we all owe Barkeep a debt of thanks for his very helpful guidance here. Thanks! I also think it's good that we seem to be having a growing consensus that it's a good thing to present editors in the RfC with specific examples to evaluate, as opposed to asking about generalities. And I do think that we are making genuine progress. For me, that's the good news.
Now for the... you know what. As much as we really are moving forward quite well, and despite the fact that we are approaching our self-imposed deadlines, I think that we have yet to resolve some really important issues for the RfC, and we cannot just wish them away. I'd rather get it right, than get it fast. We've been saying that G/H/I/J are getting close to what we want. But I want to be honest about that: I'm not really seeing a consensus that we are there yet.
I said something earlier, and I'm going to repeat it because some editors agreed with it then and I don't think anyone has really objected to it. I've said that the RfC needs to address two issues, both of them in a way that will lead to a clear consensus, one way or the other:
Those are not proposed questions. They are issues that the community needs to answer, in a way that no one will be able to argue against once consensus has been achieved. Even if we get a consensus that, no, we should not be citing drug prices to those sources, we still need a further consensus that, yes, we should present the information this other way. That's important: we need to get consensus for something and not only against something else. And I've also said that I don't think that we can really accomplish that in a single question. I also don't want to leave the RfC format so open that we fail to get focus in the responses.
So: it seems to me that we need to think outside the box, compared with G/H/I/J. I've been thinking about this hard, and it seems to me that we need to present the community with two or more specific choices. For each of those two or more, there should be specific examples of what it would look like on the page, how it would be sourced, and the policy-based rationales for it (or against something else).
If editors here are receptive to that, I can propose what it might look like, but I don't want to do that unless there actually is interest. -- Tryptofish ( talk) 22:16, 1 January 2020 (UTC)
"The wholesale cost in the developing world is about US$27.77 per month as of 2014"
"In 2014, a non-profit organization sold 250 mg tablets of ethosuximide for US$0.1845 each to recognized healthcare organizations in the Democratic Republic of the Congo"
Dear friends, I am unwatching this page because, IMO, our considerable efforts here are being undermined by back-channel conversations, and the topic of drug pricing has not been confined to this remit as required at ANI. A few of us are doing all the work to solve problems that aren't being addressed as the ANI receommended. Sorry, bye. Ping me if there is any urgent need for my useless and verbose opinions. I will keep the RFC formulation page started by WAID watchlisted. Regards, SandyGeorgia ( Talk) 14:18, 2 January 2020 (UTC)
Do you think that this content complies with Wikipedia's standards for verifiability, due weight, no original research, what Wikipedia is not, and how to write a lead sectionis biased. It is the kind of question we ask of content every day. It is a very ordinary question for Wikipedians. It is really very neutral. Are the examples neutral? Well look at User:Colin/MSHData and see if you think they are representative. I think (with the substitution of diazepam for mebendazole) they are. If you think the background and info is biased then please argue specific problems. It is also asking for comments, not a vote, which is something I very very much support. All wiki wisdom suggests a plain vote on polarised options (which is your RFC) is a recipe for polarised and adversarial comments and disunity and in the end "consensus" by counting votes. I have no doubt that some in the pro-prices faction will totally ignore the factual, source and policy issues, and present their case based on Big Pharma suppressing prices and Wikipedia being Not Censored. We will see sweeping statements that of course WHO/MSH and Medicaid and BNF are totally reliable secondary sources. And drug pricing is such an obvious world concern you'd have to be a drug company shill to want to censor prices. And some people will be totally cool with wiki-docs doing original research. This will happen. Let it happen. If Wikipedia decides those things are more important then so be it. -- Colin° Talk 22:16, 4 January 2020 (UTC)
Thanks for the ping, Tryptofish, because I did miss that.
I've been one of the regulars at WP:RFC for about a decade now. I consider accusations of a "biased" question to be par for the course on contentious subjects, and if you want to search the old archives, you will find that my opinion is consistently that claims that "The question is biased!" mean "My side is losing!" I'm not afraid of seeing those accusations, and my plan is to ignore them, because they say more about the fears (and usually inexperience) of the accuser than about the question. (Now, if you personally thought that the question was biased, I'd be very concerned, but that doesn't exactly seem to be your concern.) If you'd like, we could ask the other RFC regulars whether they think the question is biased.
The other theme in your comments is that the results might be hard to interpret. I agree with you: That's a risk. However, I see this phase as information-gathering, and then (as stated in the RFC "question"), I expect the next phase to involve a proposal that people can be for or against.
The way I expect to handle this is to care less during this RFC about summative "votes", and more about the rationales. Let's say, hypothetically, that editors read the three RFC examples and say, "Ugh, that's all wrong". What actually matters to me is the next sentence: Do they think it's bad because per-pill costs were converted to monthly costs? Or is that okay, and the problem is that the sentences are in the lead but not in the body? Or because MEDRS suggests a five-year timeline for sources (all the WT:MED regulars know what I think about that) and 2014 prices are now technically six years old? Or because they think NOTPRICE for a drug requires a source saying that the price of specifically ethosuximide itself is terribly important to the world, and no amount of sources saying that the price of WHO Essential Medicines for epilepsy can possibly suffice to prove that the price of ethosuximide is worth mentioning? Once we've seen which points of policy and common sense people care about, I think we can build on your draft to make an actual proposal for adding a section about prices to MEDMOS. Or, to put it another way, the purpose of this RFC is to make yours produce a clear consensus for fixing this guideline. This is just the necessary baby step to get us on that path. WhatamIdoing ( talk) 01:52, 6 January 2020 (UTC)
;-)
I have more than once attempted to shorten something and ended up with a significant increase in the length. I can simplify, but shortening is not my strength.
Tryptofish I appreciate that asking only about MSH-sourced text will not reliably put out the fires at NADAC or BNF or Tarascon or Drugs.com sourced texts. But if anything can be seen from the volumes I've written about the flaws in our texts, is that it is amazing that one can make so many mistakes and commit so many policy crimes in just a few words. I think if we try to explain why the "drug prices everywhere" approach is not just not-encyclopaedic but also totally impractical, for all examples, we will get totally bogged down. Fatigue will then likely to see over-simplistic replies like "Support: we using secondary and government sources for prices such as Medicaid, MSH, and Tarascon". We need to allow participants a chance to focus on a smaller level of practical issues than "all drug price sources".
I think Wiki largely discourages "prices everywhere" on fundamentally practical grounds, more than on "encyclopaedic" grounds. We aren't a standard paper encyclopaedia and already contain much trivia or dubious lists of facts. Practically, there are plenty sites (GoodRX, Drugs.com, BNF in the UK) where patients and readers can reliably find out about drug prices in their country, and get figures that are accurate TODAY for the indication or prescription they are concerned about, rather than citing a book from 2015 for a totally unknown indication and dose. All the evidence suggest Wikipedia is crap at this. Same goes for the prices of other things from mobile phones to insurance to properties in your area to train tickets.
I think it is a huge mistake to frame drug prices round concepts like primary and secondary sources, which Wikipedia has historically had mixed interpretations of meaning. James says above that the BNF is a secondary source, and it is true that it gets its information from the NHS, who in turn give a mix of regulated price and/or indicative price for the drugs, the latter of which comes from drug companies and pharmacy contracts. A better distinction is that all those sources (with the exception of Tarascon's $$$ symbolic prices, which have their own huge problems) are simply databases of prices of products with barcodes, and all those sources give their own unique kind of price (all different variants of wholesale or retail, some actual, some list, etc). Those sources are raw data, at a level of multiplicity and complexity that none of us are proposing is reproduced on Wikipedia.
Wrt notability of prices, saying "secondary sources have written about issues that are specifically about the pricing of those medications". Repeatedly it has been claimed that the fact that e.g. BNF include prices for all their drugs (similar for Drugs.com and Tarscon's book) means they have been written about. Clearly the authors of those sites/books, when writing about each drug, considered it relevant to give price data. And Google will find someone somewhere mentioning that X is a low cost drug compared to the new drug Y which is expensive. But doh!, all new on-patent drugs are expensive and most existing generic drugs are cheap, so that isn't exactly news to anyone that someone might make that remark in print or online. The advocacy argument for including prices is so strongly held by some, that we need a much higher barrier-to-entry than simply being written about in secondary sources.
I'm very nervous about get-out-of-jail clauses like "or used only with care". We aren't here to redefine fundamental policy. Anyone can argue they are careful. We see in the lead RFC that e.g citation excess is justified on the grounds that there is no policy against citations: any guideline-caution or recommendation to seek per-article consensus about citations is simply cast aside. An "or used only with care" clause simply says one can ignore the preceding text.
So I don't think, sorry, your RFC is appropriate today. Let's start with the RFC on MSH-sourced price statements today and see where that takes us. We can learn lessons from it. It may be that Wiki so clearly rejects raw-database-sourced drug prices that existing policy on WP:NOPRICES becomes the clear consensus, and we all do already know how to write about prices when newspapers, etc have made comments about them and give us a price-to-treat or a price-per-year without us having to get our calculators out. -- Colin° Talk 08:54, 3 January 2020 (UTC)
If there are any "back door" discussions, I'd like to know where they are. ... --Tryptofish (talk) 23:36, 2 January 2020 (UTC)Putting this here per your request, SandyGeorgia ( Talk) 23:38, 3 January 2020 (UTC)
User:Tryptofish thank you for the notification at WP:MED. I was not following the debate but would like to participate as a newbie. As I understand it, there area two parts of the debate. The first is regarding point #5 of the WP:NOTDIRECTORY which states,
"Sales catalogues. An article should not include product pricing or availability information unless there is an independent source and a justified reason for the mention. Encyclopedic significance may be indicated if mainstream media sources (not just product reviews) provide commentary on these details instead of just passing mention. Prices and product availability can vary widely from place to place and over time. Wikipedia is not a price comparison service to compare the prices of competing products, or the prices and availability of a single product from different vendors or retailers."
From what I can read, part of the RfC will debate whether or not medication prices falls into the exceptions listed above (e.g. is there a justified reason). The remaining question will debate which medications and which sources are acceptable? Can you confirm if I have this correct and if so, is the draft on your talk page now? Also which opinions you're seeking at this time. Thank you for shepherding this topic. Ian Furst ( talk) 15:06, 4 January 2020 (UTC)
User:Barkeep49, I think Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices is ready. As I understand it, ANI thought it would be best if an uninvolved admin endorsed its neutrality. Can you post a request at some suitable forum to find volunteer for that step?
Here are my thoughts about how to manage the RFC once it's underway, and I'd like to hear all of yours, too.
First, I'd like to encourage all the "principals" in the original dispute to stand down for the first day or two. Let's imagine that an admin certified it as neutral very soon. In that case, I could probably take the draft tag off and list it as an RFC as early as Tuesday. If that's the schedule we end up on, and if you think that other people might think you've got a dog in this fight, then please stand down until at least Wednesday. There's nothing like long or angry posts, especially from our recognizable community leaders, to scare off some contributors. I may not post my own views at all, and I certainly don't plan to do so during the early days. We can let this run for weeks, or even months if we're still getting good responses. There's no magic timer for RFCs. I promise that you will have a chance to have your say, even if your post is #10 instead of #1.
Second, I want us to be encouraging the uninvolved editors to engage in this RFC enough to tell us what they think. Some people will just want to dump a drive-by vote on the page, but if they're willing to explain their thought process, then I want to find out more. I am discouraging straw-poll or "survey" approaches, and I hope that you can all support that in practical ways, like breaking long discussions into sections with useful names like === Thoughts on X ===. Getting detailed explanations from our volunteer editors is a gift that we should treasure. I expect to be asking some editors questions to encourage discussion. If you think that you can ask a question that will draw out more details from an editor or that will encourage that editor to connect with another editor, then please consider doing that. Something like "Do you feel like your idea relates to what User:Example was saying last week?" or "Do you think that might work better with <this slight change>?" could be good. The goal is to get the other guy talking. A good, responsive question, phrased with respect, can be an excellent tool for producing further explanation. An amazing success looks like a couple of editors putting their views together to come up with something that's better than what any of them started with.
Third, if you see an opportunity to meatball:DefendEachOther, especially if it's someone from the other "side", please do so as quickly and as gently as you can. Nobody involved in constructing this RFC wants Wikipedia to get worse. We all have the same ultimate goal. If you'd like, I can ask the WT:RFC regulars to help out with this.
Fourth, I personally don't feel like I'll need an official "closing statement" to know what I've learned from this RFC. However, if you do, then please be bold and speak up now, especially if you'd like to have a "team of three" approach. Recruiting three people after an RFC has ended can be difficult and result in needless delays.
If others have advice they'd like to add, or would like to suggest a different approach, please post here. I'll make time to check this page between meetings (probably in ~12 hours or so). WhatamIdoing ( talk) 06:41, 6 January 2020 (UTC)
"Buyers: These prices should not be used as international reference prices". WHO/HAI says:
How representative reference prices are generally depends on the number of suppliers quoting for each product.. We do not have a source saying "IDA is representative of wholesale prices in the developing world".
"in many low- and middle-income countries medicine prices are high, especially in the private sector (e.g. over 80 times an international reference price); availability can be low, particularly in the public sector (including no stocks of essential medicines); treatments are often unaffordable (e.g. requiring over 15 days’ wages to purchase 30 days’ treatment); government procurement can be inefficient (e.g. buying expensive originator brands as well as cheaper generics); mark-ups in the distribution chain can be excessive; and numerous taxes and duties are being applied to medicines"they give an example:
"The price of originator brand atenolol 50 mg tablets is over 20 times the international reference price in all the countries except India (where it is still high at 5 times the reference price) and Kazakhstan. Even the lowest-priced generic is very expensive in all countries". So basically, the wholesale price of generic medicines to the state healthcare is irrelevant if the state healthcare has no stock of that medicine and the patient has to buy a premium-brand version on the private market. And that is the norm.
Quick replies:
WhatamIdoing ( talk) 16:40, 6 January 2020 (UTC)
Just a note to all involved that I decided to take a break from this yesterday and am catching up on it today. I want to acknowledge that I've seen WAID's comment above but want to have caught up fully before I launch. As I am fairly busy at work this week, I may not be able to fully catch-up here until this evening. Just wanted to set appropriate expectations.
Barkeep49 (
talk) 17:19, 6 January 2020 (UTC
I wish to add a link to Wikipedia:Prices#Discussions_about_best_practices to the RFC on pharmaceutical drug prices. This link presents a list of all previous discussions of drug prices, which I feel match the subject of this RfC. The point of sharing links to this previous discussions would be to show the history of Wikipedia community discussion of drug prices.
SandyGeorgia objects, saying with a revert that "this is not an RFC on pricing, this is an RFC on source --> text integrity". What reason is there to avoid presenting this archival collection of previous discussions? Blue Rasberry (talk) 15:26, 7 January 2020 (UTC)
IMO, the editors representing the polar opposite lead "sides" of this discussion should refrain from directly editing the RFC. I have said this before, but now James is adding contentious statements or replacing/removing factual information with over-simplified statements. He has disputed that prices are "incorrect" and has replace the
with
The problem with, what WHO admit is simply a "basic definition" is that it is over-simplistic and misses out the key factor of DDD that our article gets right: it is a statistical measure of drug consumption, a rough estimate for population-level studies. This statement implies this might be an average Prescribed daily dose and it really really isn't. Let's look at what WHO say:
So in practice, the figure is much more complex and involves a human making a non-clincical judgement. Hence the original text said "a complex statistical concept".
The
WHO also say: "The main purpose of the ATC/DDD system is as a tool for presenting drug utilization statistics with the aim of improving drug use. This is the purpose for which the system was developed and it is with this purpose in mind that all decisions about ATC/DDD classification are made. Consequently, using the system for other purposes can be inappropriate."
. The DDD is also not updated to reflect changes in prescribing practice or recommendations. A better definition would be (from WHO)
So I think we should stick to linking to the wiki article, and reverting back to the original text which made a neutral statement about this technical metric rather than offering an over-simplistic statement that will mislead. Our readers who will wrongly think that "average maintenance dose per day" reflects some clinical-practice average prescribing statistics rather than some pen-pushing statistician looking at drug dose remarks in the BNF or wherever, and choosing either the initial, the max dose or the mid-point between these depending on the words they read. It is simply "a complex statistical concept", a "technical metric", with no clinical application whatsoever, and we should not pretend it is in any way related to actual clinical practice.
James has repeatedly stated a view of DDD and how it can be used that is at odds with what WHO use it for and that WHO firmly discourage. For that reason, he should not be the editor to make claims about it in the RFC. He is welcome to state his personal beliefs in the response to the RFC.
Wrt the "Many of these are from 2014 or 2015 or are otherwise incorrect" tagged as "dubious", we have megabytes of prose outlining how incorrect these prices have been. Not just in ways that editors may disagree about but simple maths and statistics errors. I don't think that is dubious at all. Below we had a discussion about Carvedilol where for over a year we have incorrectly stated the "wholesale cost per dose is less than 0.05 USD". If fact there carvedilol ER tablets costing $6.44, $6.61, $7.08 and $6.57 for each 10, 20, 40 and 80mg doses. -- Colin° Talk 13:21, 9 January 2020 (UTC)
"The MSH International Medical Products Price Guide is a generally unreliable source of information about drug prices, because WHO only regard an "international reference price" as being "representative" if it has many supplier records, and many drug records in this database lack any supplier records and many others only have one or two". But we're not doing that approach in the RFC and I'm not about to start edit warring with others to push for that in the Background section. James has contentious views on what DDD is and how it can be used, and this is at the heart of the dispute. Therefore he should not be permitted to write about DDD in the RFC in a way that tends to support his views. If you value his and my opinion, you can wait for that in the response section of the RFC. Does that not seem reasonable? -- Colin° Talk 15:30, 9 January 2020 (UTC)
James (can someone else please ping him) you are upset about the "Many of these are from 2014 or 2015 or are otherwise incorrect". I am also upset that DDD is not currently being described as "a technical drug use metric" which is also a definition from WHO, and very much gets to the heart of its primary purpose. Instead we use a definition which, without supporting explanation, looks like the definition of average Prescribed daily dose or Recorded Daily Dose, and it very much is not. So I propose a compromise. If you will accept changing:
the defined daily dose ( "the assumed average maintenance dose per day for a drug used for its main indication in adults"; not necessarily the dose any person takes, especially when the same drug is prescribed for multiple medical conditions).
to
the defined daily dose (a technical drug use metric)
Then I'll not complain if "Many of these are from 2014 or 2015 or are otherwise incorrect" gets dropped. Our readers can work that much out for themselves. Obviously WhatamIdoing and others may have their own opinion on acceptable compromises. -- Colin° Talk 10:33, 11 January 2020 (UTC)
The RFC assertion that many prices are "incorrect" has been challenged by Doc James and AlmostFrancis. Of course, IMO the prices are incorrect on so many levels it is hard to know where to begin. But apparently even high-school rules of statistics are open to question on WP:MED so I'll stick to listing below mistakes of basic maths or where unsourced statements about "dose" are clearly wrong per reliable sources. And I'll stick to using the same methodology wrt cost per DDD that James has used (which itself is contestable). I've only picked 6 examples from the 31 drugs with the letter A that cite MSH (322 drugs total). So that's one in five with clear mistakes:
-- Colin° Talk 18:33, 11 January 2020 (UTC)
The original RFC assertion that many prices are "out of date" and even now mentioning they are from 2014/2015 has been challenged as biased by Doc James and AlmostFrancis. I have compared the unit prices in the MSH price guide cited for 2014 and compared with 2009, five years previously. This can give us an indication perhaps of how much prices change in five years since 2014 data. As above, I'm just looking at 31 drugs starting with the letter 'A'.
So that's 50% (16 out of 31) of drugs where one of the price range figures has doubled or more, or halved or more at a five year interval. And three drugs with extraordinary shifts. Us Wikipedians should be evidence and sourced-based in our assertions and claims. Disagreeing with a statement because it doesn't suit one's case is not enough. -- Colin° Talk 18:56, 11 January 2020 (UTC)
I have also compared the prices for the following year (or in a few cases, the previous year as there was no following year data).
So that's 25% (8 out of 31) of drugs where one of the price range figures has doubled or more, or halved or more, from one year to the next. And two drugs with extraordinary shifts.
In summary, annually we see a quarter of drug prices change by a factor of 2, and over five years we see half of drug prices change by a factor of 2. Some prices will change by huge amounts. We have seen claims that drug prices don't vary that much, and this research shows conclusively that this is untrue. Further, nearly all prices change annually by an amount that makes it hard to justify quoting a price to 4 significant figures. Using the median supplier price where there are three or more suppliers would result in less variation, though that would permit only a small minority of articles to quote "developing world" prices. -- Colin° Talk 13:29, 12 January 2020 (UTC)
Why not include a pro / con statement at the end of the RfC? In the United States these statements come with government voter guides before the election. In my view, the text of the RfC is not easily understandable by anyone who is not already invested in the issue. I question whether this text could inform anyone of the complexity of the issue if they are not already informed. Elsewhere there was wish that the regular players not immediately jump into the discussion in hopes that early voters would post original comments rather than ride in support of the status quo positions.
The people who are already in this discussion have chosen the best identified arguments for presenting their perspective. In conventional debate the various sides put forth their best interpretations. Let's include positions in the RfC because that seems like the conventional way to do this. What reasons are there to not do this? Blue Rasberry (talk) 16:01, 9 January 2020 (UTC)
Ian_Furst and Bluerasberry, please carefully read WAID's carefully crafted RFC and also her recommended approach at the top of this ready-stead-go section. She's explicitly asking for new voices, new thoughts and for them to be based on looking at a small set of representative examples of article text using one database source. What I believe she isn't looking for is a dump of James and Colin's opinions and then for the community to discuss our opinions, biased on their impression of whoever may or not be a World Expert on Healthcare Information Sharing :-)!
I don't believe the RFC is hard to understand:
Main question:
Do you think that the content in the examples above complies with Wikipedia's standards for verifiability, due weight, no original research, what Wikipedia is not, and how to write a lead section?
Our policies require simply that an intelligent non-expert reader be able to look at our article text and look at the source given (and only that source) and determine that the article text is supported by the source. It is a little more work for a reader to determine WP:WEIGHT for that requires one to invest in reading the body of literature on a topic, but it isn't rocket science to consider if choosing one particular (unspecified) strength and formulation of the drug and one (unspecified) indication in order to quote a treatment price is reasonable or not. Our lead guideline also gives advice on contents of the lead vs body. We really really aren't asking folk to have to read all this lengthy debate. Only to examine as you would any wikipedia article and source. Rather than opinions, perhaps I could offer some questions?
Perhaps you can think of your own. They aren't hard questions, or ones that require hours of study or a degree in medicine. -- Colin° Talk 20:15, 9 January 2020 (UTC)
(BTW I'm not wanting answers to those questions here. That's the point of the RFC.) -- Colin° Talk 20:33, 9 January 2020 (UTC)
Do you think that the content in the examples above complies with Wikipedia's standards for verifiability, due weight, no original research, what Wikipedia is not, and how to write a lead section?
I feel the drafts are too long and complicated to get meaningful input from a broad range of editors via an RfC, so I've created a very simple version for consideration: User:Levivich/Drug prices RfC draft 3. – Leviv ich 20:32, 10 January 2020 (UTC)
I don't feel like this RfC has to singlehandedly solve the problem with drug prices
. It doesn't have to be
The RfC to End All RfCs, it doesn't have to be
One RfC To Rule Them All. It does have to be digestible by dozens of editors who know nothing about this. Instead of the heart of the dispute
, pick one important part of the dispute (lungs, brain, really any major organ will do), and resolve that. Then, pick another. Like all good RfCs, it should be phrased in such a way that when it is over, someone can make an edit (or is prohibited from making an edit), and everyone will know that this edit does (or does not) have consensus based on such-and-such RfC. –
Leviv
ich
22:48, 10 January 2020 (UTC)
(sigh). Levivich, I don't think it is helpful at this stage of a multi-month discussion on formulating the RFC, to turn back the clock and basically start again. The RFC proposed by WAID is the result of this process. We all had a chance a month a go to propose "our RFC question" and the current text came about from thinking about and discussing that for a long time. Let's not begin 2020 with "I want my question" "No, I want my question". I'm sure everyone has an opinion how they'd do it if they had their way, but please respect the process we have all been through and the result of that process is Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices. -- Colin° Talk 10:24, 11 January 2020 (UTC)
Levivich, you've put a lot of work into these drafts, but I come back to the simplest possible format put forward so far is the one put together by WAID, and proposed at AN. It is true that editors outside our realm may not engage, but I do not think your drafts spell out to them what they are engaging, or draw them in and promote interest, as WAID's version does. If we complicate too far, we get no feedback. If we simplify too much, we get useless feedback (your comparison to MEDLEAD RFC asking a simple yes/no question). I hope we can get WAID's version moving, so that we can put these source-->text issues behind us and move on to asking the real question of where the community stands overall on NOTPRICE. IF no one engages WAID's RFC, we are at least not further behind, and it is preferable to have people not engage than to have people not understand what they are engaging. I think WAID's version gets to the heart of that matter: how can these sources be used. SandyGeorgia ( Talk) 20:23, 12 January 2020 (UTC)
A problem with this so-called simple RFC is the Stranger Things factor. Anyone coming along to this will see a question like "Should articles state drug prices sourced to drug price databases such as the International Medical Products Price Guide" and immediately remember that they still have two episodes of series three of Stranger Things to watch. They have no idea what the International Medical Products Price Guide is, but the name sounds impressive so it's probably ok. Are any articles doing this at the moment, or is this a proposal to do something new? No idea. Not sure I care then. Should we source drug prices to drug price databases? Well duh, they are drug price databases. What's the issue with that? The question says "such as" but doesn't tell me what the others might be, or if they are different in some way I should care. The second question 3B immediately has me playing spot-the-difference then I notice "wholesale prices in the developing world". Ok, it's a database of wholesale prices in the developing world. What's the issue with that? And 3C has the same question but replaces "wholesale prices" with "any pricing information". But the examples are the same, all stating wholesale prices, so I'm not seeing what different "pricing information" there might be to consider, other than "wholesale prices". What's the issue with that? I'm being asked my opinion on something I don't currently have an opinion on, and don't really understand why I should care at this point. But look, someone whose name I recognise is saying that the Wikipedians who want to remove prices are helping Big Pharma suppress price information, which kills people. And Wikipedia is Not Censored. Ok I care about that, let's Stick it to The Man, yeah. "Support per ____" and fade in the Netflix logo and some 80s-style theme tune.
Compare this with Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices. Immediately we know there are articles with these prices. And I'm being asked "Do you think that this content complies with Wikipedia's standards for verifiability, due weight, no original research, what Wikipedia is not, and how to write a lead section?" Right, I'm a Wikipedian. I can do this. This is something I actually know something about and care about. I'm gently introduced to the source being considered and offered a few examples so my brain doesn't overload. I can see that using this source isn't simple [we are here for months discussing this since it isn't simple]. I don't even, at this point, need to have an opinion about drug prices, just care about Wikipedia's core policies and values. Pass me a slice of pizza, this sounds worth discussing. Netflix can wait. -- Colin° Talk 20:53, 13 January 2020 (UTC)
(outdent) If it would help, we could move the two pages AlmostFrancis complains about into either project (sub page of WP MED) or Wikipedia (sup-page of this) namespaces. The information contained is not personal in any way, and claims they are biased are founded solely on personal attacks of the creator of the pages. -- Colin° Talk 11:23, 14 January 2020 (UTC)
I note that Levivich created their replacement draft, because they felt WAID's was "too long and complicated" and that an RFC "does have to be digestible by dozens of editors who know nothing about this". Levivich wants to ask "should we use the source at all" rather than "how to use the source". But this does not get to the heart of the dispute. Nobody is saying the price guide must never be used at all (though it being a dead parrot suggests its further use is limited) and nobody is saying similar drug databases should never be used at all. I feel most of Levivich's many questions are in fact ones nobody is asking, and don't need asked. IMO the heart of the price problem isn't one of "should" but "can" and "when" and "how". Should is the weakest argument of all, and can be based on opinion and advocacy as we have seen. As I have said before, Wikipedia is full of trivia and editors will tend towards including data that somebody feels is important. If drug prices were stable, were international and were simple enough that there is one obvious treatment cost per drug, we may not be having this argument at all. The problem is practical. It is a can, when and how problem. All the questions in Levivich's ask should we do something, without getting the participants to wonder how. If we want "editors who know nothing about this" to engage we need to give them the background that WAID's does.
It is ironic that Levivich has previously stated they are aligned with WP:NOTPRICES in thinking drug prices should only be included when exceptional and yet their RFC is supported by two editors who appear to want drug prices universally. In contrast WhatamIdoing has for a long time wanted financial information in medical articles but clearly wants it done right, rather than any old way that might break policy. Rather than asking a polarising "should we do this ever" question, they way to find consensus is to agree to do it right, and on a case by case basis, doing it right might mean often avoiding one source that has limitations, or wording our articles to make those limitations clear. I urge Levivich to reconsider: your "simple RFC" will be hijacked with opinionated global price concern issues if it were to be launched, and you will not get anyone even considering the result you want. -- Colin° Talk 11:23, 14 January 2020 (UTC)
Levivich's many questions are in fact ones nobody is askingis a feature not a bug :-) I'm not sure what you mean by
I urge Levivich to reconsider. There isn't consensus for any of my drafts, so I don't see what there is to reconsider. I still think WAID's RfC fundamentally asks the wrong questions (and that was the reason for my proposing alternatives in the first place), and I still think that an RfC that solicits discussion, as opposed to an RfC that solicits a decision, will not bring us meaningfully closer to a decision. Most everyone here disagrees with me on this. The only way to find out is to run an RfC. And with all due respect, I just had this exact same conversation with WAID a few days ago, above in this very same thread, so I feel we're kind of running in circles here. My humble suggestion at this point is to move anything objectionable in WAID's RfC (the background, the links, whatever it may be) out of the RfC question and instead present that information in the first few !votes (which the editors here will undoubtedly be casting), if only so as to move the ball forward and get to a point where we launch an RfC that has consensus among editors here. I don't know whose "law" it is, but we are proving the law that as discussion continues, the probability that we lose editors to retirement or sanctions approaches 1. It's clear the cavalry is not coming, so let's address whatever issues are left with WAID's RfC and get on with it. – Leviv ich 19:23, 14 January 2020 (UTC)
Draft RFC at: Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices
I think we have reached a stage where we should now look to complete polishing Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices ready for launch. The concerns raised above are:
User:Colin, I think that the idea of "Is this encyclopedic?" would probably end up in the WP:NOT-related section, but it could also fit into the subject-matter section or possibly the style section. I'm not convinced that further divisions are really warranted unless we get a huge number of comments. WhatamIdoing ( talk) 20:28, 17 January 2020 (UTC)
An uninvolved admin has commented on the draft RfC over at AN: Special:Diff/936639234 – Leviv ich 03:11, 20 January 2020 (UTC)
"You can see several hundred examples of how drug prices are being presented"or even just " What we've got on wiki? If one of both of these feels neutral enough to you, is there a version of In the real world that could also be made neutral? Or are you suggesting the only way for the RfC to be neutral is without these sections no matter their form? Best, Barkeep49 ( talk) 00:15, 21 January 2020 (UTC)
Is this the right place to note some concerns I have about some of the phrasing in the proposed RfC question? If not, please feel free to move them to the appropriate place.
"and none are comprehensive". I find that an unsupported judgement. Databases generally have a fixed range of content, so whether or not they are comprehensive isn't an issue. They either contain reliable information or they don't. I worry that the phrase places the databases in an unduly negative light.
"Editors have raised concerns about prices being outdated or having other problems", while true, it doesn't give the reader any context. From what I can see this refers to about a dozen editors over the space of several years. What about all the editors who are not concerned? don't they get a mention? What of those who think that having the latest prices available is as good as we can get on Wikipedia. Atre their views not pertinent to the background that frames the RfC?
"standards": does Wikipedia have standards written down? I know we have policies, guidance, best practice, and advice, but I don't think we have written standards. I worry that the use of the word will create a non-neutral expectation that our guidance has to meet unrealistic standards. The whole point of supplementary guidance is that it caters for the exceptions to more general policy and guidelines. I worry that it leads the respondent to make a judgement appropriate only for general guidelines, not for subject-specific ones.
"how to write a lead section": are we going to measure potential guidelines that will apply to the whole of a medical article against a guideline that is only concerned with the lead section? Or is this RfC intended only to apply to the issue of drug prices in the lead? In that case, surely the background should clarify that arguments against drug prices in the lead should not be predicated on information in the lead not appearing in the body, as that is already covered elsewhere? -- RexxS ( talk) 17:44, 21 January 2020 (UTC)
lists drug price information for WHO Essential Medicines" according to the lead of our article. If you look at the list of essential medicines, do you see any missing from the price guide? If not, then the database covers its intended content comprehensively. I haven't found any missing, but of course, you might. Nevertheless, the concern I have is that comprehensiveness is irrelevant for a source. Either we accept the source as reliable or we don't. Being comprehensive has no bearing on that. Of course we could only use the database for drugs that it is deemed to have reliable information on, but again, comprehensiveness has no bearing on that. Apologies to all who've heard the argument before, but I was asked. Do you really want me to explain my other concerns in detail as well? -- RexxS ( talk) 21:18, 22 January 2020 (UTC)
"Editors have raised concerns about prices being outdated or having other problems". I contend that without context, it places the use of prices in an unwarranted negative light. Yes, some editors have raised concerns about some prices, but if you examine User:Colin/PriceEdits #Drugs, you'll see that the majority of prices have not had any concerns raised. Of those where concerns were raised, only one was a concern about outdated prices, although one other price was updated to a more up-to-date one. I'm concerned that the statement in the proposed RfC doesn't neutrally reflect the evidence.
User:Barkeep49, I'm still hoping to figure out what the problem with the "real world" section is. Surely it's neutral to point out, as background information, that regular strength and extra strength don't cost the same, or that many prices are confidential. The editors expressing concern about it being non-neutral have been fairly vague about what, exactly, is non-neutral about it.
RexxS, being "outdated", if you apply MEDDATE's five-year standard, affects about a third of what's in articles and about 90% of what's cited to the MSH database. The category of "other problems" includes:
So you're right when you say that "some editors have raised concerns about some prices" before this discussion started, but it would equally be true to say that some editors have raised concerns about nearly every drug price in the entire wiki during this discussion, and that some of these points are being raised by uninvolved, non-medical editors ( example). Off hand, there are probably no small-molecule WHO essential drugs whose current price content (or at least some of it) doesn't fall afoul of at least one of those complaints in some fashion.
For the record, I don't agree with all of the concerns that have been raised. Even among concerns that feel valid to me, I believe that many can be fixed without removing all of the content about prices.
Also for the record, the list of concerns and the (very approximate) percentages given above are from my memory of previous discussions. I've probably forgotten some of the concerns. Some of the numbers I give might be wrong, and others apply only to relevant subsets (e.g., MSH-citing articles, or the most recent year's date in the MSH database). But I do not think it would be fair to describe the concerns as only applying to "some" of the prices. I wonder if, having seen these estimates of how many articles could be affected, you would still describe it that way. WhatamIdoing ( talk) 01:35, 23 January 2020 (UTC)
It looks like this is truly set to go. I know SandyGeorgia has some concerns about the discussion sections. Sandy is this worth holding the launch of the RfC? I am tenatively planning on launching this in 4 hours (23:00 UTC). Best, Barkeep49 ( talk) 19:02, 23 January 2020 (UTC)
This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Archive 5 | ← | Archive 10 | Archive 11 | Archive 12 | Archive 13 | Archive 14 | Archive 15 |
I have a procedural/administrative question. I see that there is a DS notice at the RfC draft page, based on this being related to WP:MOS. It had never occurred to me that we are working under a DS situation here. Are we? -- Tryptofish ( talk) 20:20, 4 January 2020 (UTC)
There has been a fair amount of requests for clarification and interpretation occuring at my talk page (see
[1]). Out of respect for the consensus that "The question of drug pricing is remitted to a single venue"
I note the above diff and then copy over the current discussion that had been occurring there to here to be continued. Best,
Barkeep49 (
talk)
15:51, 3 January 2020 (UTC)
Content from Barkeep49's Talk Page
|
---|
The following is a closed discussion. Please do not modify it. |
Just a note to those around here, I was just asked on DocJames talk page specifically about updating. I
FYI, User:Colin/PriceEdits contains a computer-generated list of all price/cost insertion/deletion edits to 530 drug articles by any editor since 2015. It also contains my analysis of where editors have come into conflict. Plenty examples of why I note my concerns above. To take an example of the kind of "copyedit" being suggested by User:WhatamIdoing at the MEDMOS discussion, an editor changing "the wholesale price in the developing world is" to "The median buyer price according to the International Drug Price Indicator Guide was" sparked an edit war at Lactulose. -- Colin° Talk 10:40, 2 January 2020 (UTC) |
I look forward to learning about the flaw that Sandy sees with allowing the updating of information. Best, Barkeep49 ( talk) 15:51, 3 January 2020 (UTC)
I'm back; thanks for the well wishes. While I was driving, I tried to organize my thoughts to avoid my usual verbosity. It is unfortunate to clog this page with this feedback, but I understand Barkeep's reasoning for wanting to keep this here.
I would like to unpack my concerns in steps, so as not to drop a wall of text filled with my usual typos on this page, and for us to be able to calmly digest the different components.
Starting with No. 1, I felt completely betrayed yesterday. Usually I apologize for an outburst, but I hope no one feels I have something I need to apologize for, because I pushed back from the computer and hit the break button before I completely exploded. I always apologize for misunderstandings when my wording is not careful, so in that vein, I am sorry if Barkeep felt I was questioning his integrity or saying he had disrespected me or us or done something intentionally deceptive or malicious; that is not the situation.
But yesterday did not feel like this month was well spent. The ANI started on December 5, and today is January 3; we expended the entire holiday season on this, and we all typed 'til we were "blue in the face" with unfailing good faith and very little need for, as Barkeep said, people to walk things back. (I will discuss in part 4 whether some people should have been asked to walk things back, but my concern there is not aimed at Barkeep.) Those who participated here (which I will later distinguish from the MEDLEAD RFC) showed that they can address this dispute calmly and without behavioral issues, which is part of why I do not feel we are yet to the arb stage. As we have all expressed (and I believe sincerely meant), Barkeep has gone above and beyond the call of duty and has been exemplary in his treatment of all of us. Yet, through no malintent or negligence, we came yesterday to a most unsatisfactory juncture, for reasons I will unpack bit by bit here.
Basically, I feel our good faith was taken advantage of. I believe the problem yesterday happened because Barkeep has been shielded from full information of what he was walking in to, by necessity, because we are not allowed to "rehash old grievances". So, Barkeep could not have foreseen yesterday's reaction to something that, to many of us, was entirely foreseeable, because it precisely fits the pattern we have been dealing with. I'll unpack that in part 3.
When Barkeep posted to AN, I cheered him, because it has become obvious that he was working too hard. When he retracted his AN post, I worried. I find it a bit disgusting that there are probably twice as many people who participated at the ANI than have participated at the MEDLEAD RFC, and it is unconscionable that what I asked in the ANI (when I laid out my terms) that more eyes watch these events, and Barkeep closed by asking for more volunteers, he got none. I would like to revisit the ANI when I get to Part 5.
Kevin, thank you very much for stepping forward with reassuring information, but I never thought that Barkeep's conduct back-channel needed to be questioned; he has given no reason for doubts about his integrity or how he would approach other admins. My concerns had to do with him not knowing the full picture, and perhaps unwittingly falling into something he might be unaware of.
And that relates to Part 2. Barkeep, I have emailed you a description of things that have happened to me in the past on Wikipedia-- many years past-- as a female editor. To be able to move forward, I would like to ask if either you or Kevin know why the third admin on IRC has not self-identified as Kevin did. I hope you can view my question in the context of the private information I sent you, and understand that it would be reassuring to me if that admin would self-identify. If they won't, that will force me to ask the question in a less-than-desirable way. Meanwhile, I hope we'll all give a good re-read to the ANI thread, because I want to next raise some specifics about where we stand. That's all for now; Barkeep and Kevin, please let me know if it is likely that the third IRC admin will speak up. I'll continue afterwards with Part 3. Regards, SandyGeorgia ( Talk) 22:34, 3 January 2020 (UTC)
Why was updating ok with me when I'd said even adding a tag was not? In my thinking it's different because it's not adding something new and because the embargo is designed to ameliorate the conflict not preserve bad information.And Kevin (aka L235 said:
My reasoning was that the spirit of verifiability doesn't allow us to keep old bad information if newer, equally- or better- sourced information is available -- it must be updated or removed, and the closure explicitly prohibits removing it, so updating it is the only option. Additionally, updating pricing doesn't seem to involve the same dispute as to whether prices should be included or not.So it is easy to see that both Barkeep and Kevin are operating from a position of wanting to preserve the integrity and verifiability of information to our readers, while minimizing conflict. So far so good. Well intended, makes sense. But here's the problem. We have reams of evidence above that all of the drug price information we have presented to our readers in these 500+ examples is bogus. We have computer-generated analysis that shows essentially all of the prices were inserted by one editor. I have been asking for a month, and we have not yet seen one single example of these database sources being used correctly. So, questions are:
In the United States the wholesale cost is about US$13.50 a month. [3]
we used the World Health Organization DDD classification to analyse dose–response relationships. This classification implies that a dose of 25 mg of chlorthalidone is equivalent to a dose of 25 mg hydrochlorothiazide. This assumption is probably not valid. A recent cross-over trial suggest that chlortalidone is about 1.5–2 times as potent as hydrochlorothiazide with regard to antihypertensive efficacy.-- they admit using the DDD for a purpose that WHO explicitly warn against:
"DDDs do not necessarily reflect therapeutically equivalent doses of different drugs and therefore cannot be assumed to represent daily doses that produce similar treatment outcomes for all products within an ATC category". We are similarly abusing DDD for our cost-to-treat claims, made all the worse since there is no published information about what exact indication a DDD was calculated for.
A meta-analysis of trials of chlortalidone for high blood pressure found that lower doses of chlortalidone (e.g., 12.5 mg daily in ALLHAT study) had maximal blood pressure lowering effect and that higher doses did not lower it more. (Musini VM, Nazer M, Bassett K, Wright JM (May 2014). "Blood pressure-lowering efficacy of monotherapy with thiazide diuretics for primary hypertension". Cochrane Database Syst Rev (5): CD003824. doi: 10.1002/14651858.CD003824.pub2. PMID 24869750.)
We see a trend to greater benefit with 25mg, who is "we" and what do you mean? SandyGeorgia ( Talk) 11:59, 6 January 2020 (UTC)
A meta-analysis of trials of chlortalidone for high blood pressure found that lower doses of chlortalidone (e.g., 12.5 mg daily in ALLHAT study) had maximal blood pressure lowering effect and that higher doses did not lower it more.SandyGeorgia ( Talk) 12:16, 6 January 2020 (UTC)
Part 4: Barkeep49 said:
In the time that I've been working on this conflict I've worked hard to treat all editors with respect. ... But I also remember acutely what it's like when you don't have the sysop flag. I wrote, in a line I had to remove from my ACE statement due to space constraints, "I remember what it’s like to feel put down not or otherwise dismissed because" I wasn't a sysop. We lose so many good editors for so many reasons and I'm sorry that the conditions here are such that we're going to (potentially) lose you. The places you've chosen to contribute in this dispute have been made better because of your contributions. I can only hope you decide Wikipedia remains worth it.
First, thank you Barkeep for the kindness in your response. As to whether Wikipedia remains worth it, I am first and foremost a medical editor, and next, was highly involved in the featured article process. Obviously, I would prefer to contribute medical FAs to Wikipedia, and I can better spend my time IRL if I can't do what I do best here. That's why I'm here, trying to resolve this conflict.
I do believe you have treated everyone equally, and with respect. I'm glad you raised the memory of what it feels like to be treated lesser when you don't have a sysop flag. In my case, I never wanted it, not only because it would be a distraction from contributing content, but because of one of my earliest experiences on Wikipedia. I was attacked by someone claiming to represent "we admins"; [7] the position expressed there was completely fictitious, against every behavioral policy, and I had no recourse but to sit on it. A few months later, Raul appointed me FAC delegate, and a year later, that admin was desysopped. I understood then that a fact of Wikipedia's dispute resolution processes was that it takes a long arbcase and a lot of different instances of misuse of the tools to deal with admin abuse, so we regular editors had best simply accept and live with that reality.
In this instance, although you have treated everyone equally, and "adminning" the MEDLEAD RFC was not your remit (and no other admin came to your assistance), let's look at how non-sysops vs sysops have fared.
Yes, you have treated everyone equally and respectfully. But nonetheless, not everyone ends up being treated equally. I have a tough skin because of years of corralling cats at FAC, and having to stay above it and stay neutral in disputes no matter what was thrown at me. But the effect piles up, and then it is best to take a break for tea, and I'm sorry that occurred on your watch, which has been fair.
The reason it occurred when it did is because I so abhor the effects of backchanneling (something that was well understood during my tenure at FAC, and everyone knew if they cooked up support off-Wiki, I'd shut down their FAC ... and don't even think of emailing me about your FAC unless it is something like a serious COI, which did happen once, and I wrote the arbs). My issue is that if you had had the same conversation with those admins in public view, others would have had the opportunity to point out the items in Point 3, without us coming to this level of awkwardness after the fact. All of this thread could have been avoided; transparency works. I still feel awkward about putting this on this page, but understand why you wanted it here.
I believe Part 5 is now completely summarized back on your talk, but having lost a bunch of this post, I may need to go back and re-read everything to see what I have not addressed.
I think we have a way forward, in addition to the RFC, that doesn't require arb intervention at this point. I believe it has now been revealed that the premise that the ANI close was founded upon was faulty, and we can revisit. That does not mean we should not launch the RFC that we have worked so hard on. Best regards, SandyGeorgia ( Talk) 22:57, 4 January 2020 (UTC)
"We have reams of evidence above that all of the drug price information we have presented to our readers in these 500+ examples is bogus...[removal of conduct related discussion addressed on my talk page]... I have been asking for a month, and we have not yet seen one single example of these database sources being used correctly."This is where I think we've been ill served that the people who feel that pricing information has been used correctly are not participating as actively as those who oppose it. Though I will note that at times both WAID and Trypto have presented modified versions that they feel is policy compliant. Those examples have not been satisfactory to you which is fine. Different Wikipedians can have different interpretations of our policies and guidelines. So I would suggest in fairness that your second sentence is really "we have not yet seen one single example of these database sources being used correctly in my expert opinion" (with the words in bold being my addition). This kind of situation is what the essay WP:POLSILENCE is talking about because everyone reading this should know that the silence here doesn't mean that everyone agrees. And at the end of all of this, that's why we're having an RfC to find out if people agree with you (and the interpretation of policy and guidelines that you and others who share your beliefs cite).That RfC, hopefully, gets us a consensus that lets us start to adjust articles around drug pricing again. In the interim when asking to interpret the embargo I have been attempting to balance "What will keep this conflict from escalating?" with "What will serve our readers?". So adding disputed tags serves our readers but did not, in my evaluation, do enough to serve our readers to outweigh the potential that had to escalate the conflict. Similarly adding to a draft (which obviously won't be seen by most readers) does not offer enough benefit to our readers to outweigh the potential to escalate the conflict. In this case I made the judgement that it could potentially help the conflict (by possibly showing a formulation that some editors would find appealing) while also helping our readers. Of course it could also exacerbate the conflict. That's why I decided this was the time I needed others thinking. Through a mistake filled route we got here (which I've previously addressed and accepted responsibility for) we do have a second uninvolved sysop ( L235) offering their related reasoning that reached the same conclusion. I have, for reasons I explained in more depth in the conduct response on my user talk page, not felt that any history was appropriate to consider in light of the ANI's consensus that there would be
"no rehashing of grievances.". What I read you writing here is that you weigh the competing priorities of service to readers and conflict escalation differently. Which I understand and have and will continue to consider but also does not at this point change my interpretation of that balance. Best, Barkeep49 ( talk) 04:07, 5 January 2020 (UTC)
I have been asking for a month, and we have not yet seen one single example of these database sources being used correctly.That doesn't mean there may not be one; yes, we are hamstrung because of silence of those supporting this data source. Re WP:POLSILENCE, I am not convinced that essay applies here, and I wouldn't even use it at the MEDLEAD RFC, where it might apply-- it's an essay. More significantly and to matters at hand, if we can get the chlorthalidone example moved to its own section, we can continue to explore now one example. SandyGeorgia ( Talk) 17:55, 5 January 2020 (UTC)
Collapsing not that important comment as suggested by Barkeep49.
Nil Einne (
talk)
14:17, 6 January 2020 (UTC)
|
---|
|
Draft RFC at: Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices
Thanks for getting this moving, WAID. I am unclear if we are actually restricted to generics? Why do our questions not apply to all drugs? SandyGeorgia ( Talk) 03:08, 2 January 2020 (UTC)
Colin might want to switch out the sample drugs to include one where it's not even clear what the dosage is or what is being treated. SandyGeorgia ( Talk) 03:10, 2 January 2020 (UTC)
WhatamIdoing, I think there is an initial problem with the "what the source says". We need to back up a bit to see what the source says about the drug, and not just one variant tablet size or formulation: Ethosuximide, Carbamazepine and Mebendazole. We can see from that:
So we need to be up-front that the source offered options, and the editor chose one of them. We need to be careful the examples don't offer illusory easy answers to picking one of them (such as, there's one variant that has lots of suppliers and the others have none, or that the 100mg and 200mg tablets work out the same price by dose anyway). The existing method for choosing a variant isn't foolproof: the diazepam article picked the wrong tablet with no suppliers whereas a different tablet size has lots of suppliers. -- Colin° Talk 11:42, 2 January 2020 (UTC)
[snip]
We are giving the prices and DDD but not actually helping readers with the maths. This matters because sometimes the DDD is being used and sometimes not. With Ethosuximide, if we multiply the 0.1845 250mg by 5 (1250mg DDD) and by 30 we get £27.67 (not sure where 27.77 comes from). With Carbamazepine, we are multiplying these 200mg tablet prices by 5 to get 1000mg and using that for the daily dose. With Mebendazole, we are actually just giving the price of the 100mg tablet, not the DDD of 200mg. So what do we even mean by "dose"? Since the DDD isn't being used here, it may be worth me explaining to you guys using the BNF that you can't read in the USA, which likely focuses on the kinds of parasites we get in the UK
So a "dose" could be the 100mg one-off dose, or the 200mg daily dose for three days, or the 500mg dose you take once. How on earth do we cover this? For many youngsters on Wiki, they may only be familiar with ibuprofen tablets and the contraceptive pill, and unaware that medicine dosage and indications for drugs are complex. Maybe we need a little side-box for each drug, that explains things that Wikipedia generally is forbidden to cover like how the dose is recommended for each indication/patient-group. -- Colin° Talk 11:58, 2 January 2020 (UTC)
Wrt wording "One organization said that they sold ... in 2014" should really be "One organization said that they sell... in 2014". We have price data but no evidence they actually sold any. It might be simpler to call these "organizations" "suppliers", especially as that's what the source calls them, and we will end up discussing buyers and suppliers. -- Colin° Talk 11:42, 2 January 2020 (UTC)
In the Carbamazepine example, you have focused on the suppliers. In fact, the highest price in the article text ($0.24 per day) is taken from one Buyer (SICA: System of Central American Integration). So, to discuss what's gone on in that article text, we need to talk about Suppliers and Buyers, and should really enlighten readers that for example WHO encourage we take the median supplier price, and only consider this representative of an international price if there are many suppliers. We can source this and if necessary quote verbatim. --
Colin°
Talk
11:42, 2 January 2020 (UTC)
In the Mebendazole example, the highest price in the article text ($0.04 per dose), comes from one Buyer (South Africa Department of Health) and in a package of 6 pills, not 1000. -- Colin° Talk 11:42, 2 January 2020 (UTC)
The bits at Additional information can be found at... and Previous discussions on this subject include... still need to be finished. Anyone's welcome to add whatever they want there (or post it here, and I'll add it). I'd really appreciate some help with finding all the relevant things. WhatamIdoing ( talk) 22:27, 2 January 2020 (UTC)
I don't want to be here, but here I am. Please feel free to ignore me, but. I still think there is no need to present three examples from the same database. We will get a result about one database. We could present three examples from the three different sources that have been incorrectly used, and get a broader result. It is the same principle; we have no good data on drug prices anywhere. SandyGeorgia ( Talk) 12:30, 4 January 2020 (UTC)
After a couple days away, I just looked at the RFC page, and it looks ready for launch to me. What work is remaining? Nice work, WAID. SandyGeorgia ( Talk) 18:06, 4 January 2020 (UTC)
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Actually, I'm feeling ignored, too. I think it will be a fiasco if that version of the RfC gets launched. And I've been saying so quite clearly for a long time now. If the consensus really is to do it that way, I won't stand in the way but I also won't refrain from expressing my concerns. I feel like a very small number of editors are, in effect, trying to run out the clock, and aren't really engaging with my expressed concerns, even though I've been revising my suggested version of the format in response to their concerns.
So here is what I would like. I hope that editors who have been watching here, or who have just started watching here, will provide additional opinions about the two draft versions: Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices and User:Tryptofish/Drug prices RfC draft 2.
I'm not asking for editors who have already expressed an opinion to repeat themselves here. I'm hoping that more editors will express their opinions, giving a wider range of input. Thanks. -- Tryptofish ( talk) 20:52, 4 January 2020 (UTC) A part of it struck. -- Tryptofish ( talk) 23:04, 4 January 2020 (UTC)
User:Barkeep49, others, I think things have gone a bit off the track here. This top-level section heading was created to discuss WAID's draft and is now discussing the genesis of and participation at Tryptofish's draft. The former is imo nearly ready to roll and the latter little more than a sketch of an idea (sorry, but true). After almost a month since the ANI was closed and we were asked to form an RFC, it seems we are being dragged back to square one, still scrabbling around to find even one example to put into this second rival draft RFC. It seems we have two styles of RFC:
The point of having the RFC is that we do not claim to know community consensus on this matter, and since polling is not a substitute for discussion, it seems wrong at this point to go for a simple poll, and to offer a false dichotomy that voters must pick positions on. Polling is known to polarise debate and separate voters into adversarial factions, so this would not seem to be a wise option when we have already seen incendiary claims of collusion with Big Pharma to censor Wikipedia and conceal prices from patients. We don't need closed-minds formulating some sound-bite that will destroy the opposition's argument. We need open minds to look honestly and carefully at this complex issue. So complex, that I think we do need to concentrate for now on one source and one region (developing world). I keep saying, please lets choose the least-conflict option, and an RFC that divides Wikipedians into two camps to try to outvote each other is not that, imvho. -- Colin° Talk 21:43, 5 January 2020 (UTC)
Just to make it clear, there was not an "ANI consensus [] that [Colin] should not be the person to formulate the RfC".This is correct. There was one comment in the ANI:
There may very well be problems, but I'm unconvinced that Colin is the best person to be the primary one drafting an RfC to deal with them. Nil Einne; I believe that Colin has acted correctly in respecting the spirit of that one comment, and leaving the actual drafting to other parties. I have full confidence in WAID because of her RFC experience and long-evidenced neutrality in all matters WPMED. She has done a fine job in laying out one part of the problem that needs to be examined before we can move to the broader issue of what to include in MEDMOS, while not giving respondents too much to deal with in one shot. Trypto, I hope you will back off on the term "laughable" and that we can move forward step by step, which is something we've discussed many times on this page. SandyGeorgia ( Talk) 11:15, 6 January 2020 (UTC)
Let me be clear: I am not going to write any RFC, and I don't think James should write one either.: [13] And then you told me just above that
the ANI indicated Colin shouldn't do the drafting, which I see now that you have struck: [14]. Once you told me that, I figured that I just hadn't remembered it. That's what it came from. -- Tryptofish ( talk) 20:49, 6 January 2020 (UTC)
;-)
The narrower RFC that I have drafted at
Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices was not originally my idea, but I no longer remember who first mentioned it.
|
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Barkeep49 I know you said that certain concerns should be done on a user talk page rather than a guideline talk page, but I'm prevented from doing that. I have to say I'm finding these allegations by Tryptofish to be rather against our assume good faith policy, and not conducive to collaborative working. Trypto, here you are frustrated that I haven't helped fill in your skeleton RFC and suggest I'm deliberately leaving it bare in order to comment that it is unfinished. The allegation that folk are "rigging the RFC to get a desired result", and query to WAID about her response to a potential complaint raised "that the RfC is hopelessly biased and should be discarded" do not make me feel comfortable. Your "let's not let the facts get in the way" is fightin' talk, Trypto, please tone it down. Some of us have been at this since October! Let me be clear: I am not going to write any RFC, and I don't think James should write one either. I can express my opinions about the content/focus of the draft RFCs and you guys can accept or reject it same as with any other editor here. I think we are all rather tired, and frustrated at the lack of participation by experienced editors in helping drafting any RFC, but lets please collaborate and compromise, not fight among ourselves. -- Colin° Talk 22:45, 4 January 2020 (UTC)
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Barkeep's Background
|
---|
A few days ago, Colin quite reasonably asked me for some examples of good RfCs to use as a model. I spent some time today looking into past MOS (and related) type RfCs and also asked a few people with experience closing Wikipedia related RfCs for examples. I am reluctant to share any because I don't know that they will actually help move the conversation forward. What seems to happen is that someone puts forth an idea and it gets criticized for either structure or wording. Much/all of the criticism is fair (and I've done it) but ultimately it means we aren't getting any closer to an RfC. Above Colin expressed frustration with the status quo of the information being included in hundreds of articles at the moment. A few others seem to be feeling this as well. Others are no doubt being frustrated at the inability to add the information in new places. The way past this is to get this RfC completed. Editors of all stripes need to have some faith in our process. The RfC question itself is not the right place to put forward compelling arguments about why pricing is/isn't appropriate. Instead that can happen during the RfC or in some sort of supporting material. If the goal is for the RfC is to solve every pricing related question the whole RfC is going to fail and nothing will be decided. Instead the decision should be made, by the people here who care most, about what's most important. To everyone I remind it is possible that not everything that's important will get decided by this RfC. So what's most important to decide? Some willingness to to accept that important issues won't be decided right away needs to be tolerated in order for there to be any chance of moving this forward and it seems clear that this needs to move forward. If a person's answer to "what weaknesses/compromises can I live with?" is nothing then that person is not going to be able to successfully participate in formulating this RfC.The good news is that no one seems to yet be at that point. The further good news is that multiple people are seeing progress being made. In rereading this talk page it seems like there is some level of agreement behind a single question (that can be answered with a support or an oppose). People seem OK with coming back to the details later. This is helpful because a single question also seems most likely to achieve consensus from the community. If we can't get consensus around a single question I think it could also give us insight into what the multiple questions have to be at this point and we can, if necessary (though I think it might not be) examine different formats for multiple questions. As such I am going to propose the following. |
I am suggesting we focus, for now, on trying to perfect a single question about pricing (or whatever your preferred term is). I am also going to ask that people to follow my lead and collapse extended content, leaving only the most important information visible. Say as much as you want, but let's make it easy for people to navigate. Thoughts? Barkeep49 ( talk) 21:22, 23 December 2019 (UTC)
So far the following single questions have been proposed:
Do you think that a pharmaceutical drug has one price that can be expressed in dollars and cents, for each region such as US, UK, developing world, and this is information that Wikipedia drug articles should include and can routinely be sourced by editors without original research.
— User:Colin
Should Wikipedia articles contain information about the cost of medications?
— User:Doc James
Please take one (or both) of these and wordsmith them to your heart's content. And if you want to explain, at length, why your version is good, or issues you see with someone else's proposal, feel free but again please consider collapsing those comments. Barkeep49 ( talk) 21:22, 23 December 2019 (UTC)
SG's Background
|
---|
|
Do you think that any individual pharmaceutical drug has one price that can be expressed in a given currency, for any region such as the US, UK, or the developing world?
— User:SandyGeorgia
Followup: What kinds of sources for pharmaceutical drug prices can be cited by editors, without original research, for Wikipedia drug articles?
— User:SandyGeorgia
Thanks for all your hard work, Barkeep49; this amount of effort is beyond the call of duty. SandyGeorgia ( Talk) 01:43, 24 December 2019 (UTC)
— kashmīrī TALK 02:22, 24 December 2019 (UTC)For articles discussing medications, do you think that Wikipedia can, reliably and without original research, source medication prices as used in various regions of the world; and if found, should convert them to a common currency and include in articles?
we don't want prices in South African rands, RTGS dollars or Chinese yuans, do we?- Why not? Wikipedia is an internationally targeted encyclopedia, what currencies do we want prices in?
Inconsequential formatting change
|
---|
|
I think that the main question to be settled is how much WEIGHT do we (by default) put on sources about prices (as understood in the dollars-and-cents model, not the general how-many-people-can-afford-that sense). Does it fall into (or near) the category of basic information that User:Bluerasberry calls Wikipedia:Defining data, in which case we need to include something about it whenever reliable sources permit us to say anything at all? Or, alternatively, is this content something that we should normally not include, and only mention when we have especially good sources (e.g., multiple high-quality sources that discuss the price at length). Here are some examples that we might consider:
Subject | Source type | If we put a lot of weight on prices | If we put less weight on prices | Notes |
---|---|---|---|---|
Insulin | Many long articles in news media and academic journals, including claims of price gouging and people dying because they couldn't afford the drug | In the US, between 1987 and 2017, the wholesale price of long-acting insulin increased from US $170 to $1,400 per vial.[1] | In the US, between 1987 and 2017, the wholesale price of long-acting insulin increased from US $170 to $1,400 per vial.[1] | |
Valproate | A 2017 peer-reviewed journal article, and some data points in various databases (i.e., independent primary sources) | According to estimates published in The BMJ in 2017 for drugs on the WHO Model List of Essential Medicines, the cost of manufacturing the active ingredient in this drug in India, is approximately a couple of US cents per pill.[2] | (Nothing – this is a single primary source) | The cost of manufacturing the active ingredient is reasonably consistent worldwide. India is the biggest producer of these generic small-molecule drugs. But no retailer or consumer buys just the active ingredient. |
Denosumab | An article in a pharmacy industry magazine (independent and possibly secondary) | Shortly after its original approval in 2010, Medicaid's US average wholesale price for a 60-mg prefilled syringe of denosumab was reported at US$990 per dose, with two doses expected each year to treat osteoporosis.[3] | (Nothing here, but maybe something in the manufacturer's article) | Just one dose (of two for this drug), in just one country, at just one point in time, using just one metric (of many). |
Golodirsen | An article in a biotech business magazine reporting on an Earnings call (independent and primary) | In the days after Sarepta Therapeutics received permission from the US FDA to market the drug, the net annual cost was estimated to run around US$300,000 per treated patient, assuming the patient was a child weighing 25 kg (55 pounds).[4] | (Nothing here, but maybe something in the manufacturer's article) | No actual sales and little non-business coverage at that point, but high-cost drugs tend to attract attention, so maybe more sources would appear later, at which point it might be treated more like the Insulin example. |
Abacavir | A routine entry in a government database | According to the Drug Pricing and Payment database maintained by the US Centers for Medicare and Medicaid Services, the National Average Drug Acquisition Cost for 300 mg pills of abacavir was US$0.77 each in December 2019 in the US. | (Nothing, because it's a primary source) |
All of these examples have been mentioned in the discussions leading up to this point. If anyone feels like any of the examples are misrepresented, please let me know. WhatamIdoing ( talk) 05:41, 24 December 2019 (UTC)
Colin's Background
|
---|
I think we have rather forgotten why we are having this RFC.
I hope a pattern is emerging. We aren't having an RFC about some new or alternative idea for prices in Wikipedia articles. We are having an RFC about the actual current prices in actual hundreds of articles. We're having an RFC because of an impasse between two editors. And we're having an RFC because WP:MED has completely avoided making direct explicit criticisms of that text or of fixing any problems in the past three years. WP:MED is clearly not going to fix this and we need input from the wider community and neutral editors to contribute. We need to offer a question that directly resolves this matter, rather than creating new ones. -- Colin° Talk 10:01, 24 December 2019 (UTC)
|
I think question A (by Colin) is a necessary and sufficient condition for the current text in hundreds of drug articles to be kept. Splitting in two like Sandy proposes weakens this, especially the open question B which may not likely lead to anything other than a random mix of opinions. The problem with a "What kinds of sources" question is that it always depends what you use if for and "for pharmaceutical drug prices in drug articles" is not specific enough. Many people like to view sources as adjectives. So "MSH" or "data.medicaid.gov" or "WHO" or "BNF" will be viewed as "reliable sources" and as "secondary sources" (they aren't the primary source of their data). So I suspect we'd just get comments like "Must use high quality secondary sources like the BNF" without stopping to think that the BNF may list 30 prices from 10 manufacturers for a drug, or even working out what the two prices the BNF list for each record actually mean. I'll try to find some representative article texts later. Essentially we want an "Are you happy with this?" question around existing practice. -- Colin° Talk 10:21, 24 December 2019 (UTC)
Putting this here at the bottom, to encompass much and good feedback above. Please, people, come on ... take Barkeep's suggestion and put up concrete suggestions so we can start wordsmithing and discussing specifics. Once the proposals are up, we can see the issues and refine. I am at the limit of my wordsmithing ability, and despair has set in; length has again taken over this discussion, and we have nothing concrete. Trypto and Peter and Ronz, give it a go even if you aren't yet fully satisfied with what you might intially propose. SandyGeorgia ( Talk) 19:02, 25 December 2019 (UTC)
An article should not include product pricing or availability information unless there is an independent source and a justified reason for the mention. Encyclopedic significance may be indicated if mainstream media sources (not just product reviews) provide commentary on these details instead of just passing mention. Prices and product availability can vary widely from place to place and over time. Wikipedia is not a price comparison service to compare the prices of competing products, or the prices and availability of a single product from different vendors or retailers
"And I agree with that. All the sources being used to discuss prices are independent. And we have sources that justify the importance of pricing information". We have some editors who think prices concerns in national newspapers is what counts as independent discussion and WP:WEIGHT to include. Others claim that multiple drug databases listing prices satisfies WP:WEIGHT and finding some obscure internal memo on the price of X is sufficient discussion, or that general concern about drug pricing is enough to justify the inclusion in every single drug article. So we need a question that cuts through all that to make a clear consensus. If we just end up with a reworded WP:NOTPRICES that is immediately ignored/evaded we haven't achieved anything.
Should reliably sourced and unambiguous dosage prices be routinely presented in articles on drugs (medications) in a prominent position (lead or infobox)
— User:Pbsouthwood
"The wholesale cost in the developing world is about US$0.01 per dose as of 2014.". Again there are a whole host of problems with this because the source lists only the Buyer price in the Dominican Republic and in Peru, and lists no suppliers at all. This should have run alarm bells for such a huge drug. In fact the 10mg tablet would appear to be uncommonly used (hence no suppliers at all, and in 2015 only Peru was a Buyer) and the 5mg tablet here with eight suppliers is far more reasonable. But what is a "dose". Would someone writing "costs ___ for a 5mg dose" and citing this be "reliably sourced and unambiguous" in your book? The source does not indicate which tablet size to pick or what is a "dose". Nor, for other drugs, does it suggest whether to use enteric coated tablets, or suspensions. The MSH is a "reliable source" for some things (nobody is doubting it is generally likely to be correct about the prices it lists) but not a "good source" for others (the Buyer prices in Peru and DR are not considered representative of the "developing world" according to ANY recommended usage of MSH). Other articles give a cost per day or per month or per treatment. Are they "dosage prices"? By "unambiguous" does one need to include the exact indication too. For example, the BNF gives all sorts of dose options for various indications, many in a range. How would we pick which indication? And if we picked one ("Muscle spasm of varied aetiology") the adult dose is
"2–15 mg daily in divided doses, then increased if necessary to 60 mg daily, adjusted according to response, dose only increased in spastic conditions."How does that translate to a "dosage prices" in an article? The BNF prices are here.
Would someone writing "costs ___ for a 5mg dose" and citing this be "reliably sourced and unambiguous" in your book?
I am concerned that our questions must be asked in a way that addresses the NOR aspect of these drug databases, the WEIGHT aspect of whether we should include prices at all, and the LEAD aspect of whether they should be in the lead. If we really have so many core policies being challenged, what the heck. NOR, NOT and WEIGHT are policy; why are WikiProject guidelines and practices challenging policy and why is that not being done with RFCs on the core policies? I am going to end up dissatisfied if we don't have questions that will get us to the core problems. If we still had an RFC/U process for user conduct, we would be asking these questions there. SandyGeorgia ( Talk) 14:56, 26 December 2019 (UTC)
So, let's consider the answers that F will generate, and how the closer will interpret those? F presumes a baseline understanding of and achknowledgement of reliably sourced and unambiguous dosage prices. We don't have that in this price dilemma. If we did, we wouldn't be here. So, what will the closing admin do with the "ILikeIt", "Me, too, per editor-so-and-so" responses that will not engage the core questions and policies? The RFC MEDLEAD shows we will get "because I like it" responses. We need to be highly specific in our questions about the core policies: NOR, WEIGHT, NOT. With the MEDLEAD RFC, a closing admin can argue that guidelines are flexible and can be ignored. In that case, with respondents not engaging the core questions, we end up with protracted local discussions to determine consensus on individual articles, where one group will argue LEAD and another group will argue MEDLEAD. We end up with articles that cannot be taken to FAC, because you can't please two masters. This is really not a big deal, because essentially no one at WPMED is attempting to write complete articles anymore anyway, and no one is maintaining most of the project's Featured articles.
Unlike the guideline LEAD, on core policy questions, the problem cannot be so easily overlooked. In this case, if we end up with an inconclusive RFC where respondents do not engage policy because we haven't asked the questions with great specificity, what's next? Protracted local disputes end up at arbcom. SandyGeorgia ( Talk) 16:22, 26 December 2019 (UTC)
I really think this is important: NOR is not DUE.
Deciding which entry to cite in a database is not a NOR violation. It's (possibly) a DUE violation, but an accurate description of the content published in a reliable source is never NOR. This means that if you look up wonderpam in The Database, and it says "100 mg pill – generic – Specific Price Type – US$0.10 – December 2019 – UK", then writing "According to The Database, the Specific Price Type of a 100 mg pill of generic wonderpam was US$0.10 in December 2019 in the UK" is not original research. That's actually what the published reliable source said; therefore, that's not NOR.
Now, while that statement is not a NOR violation, NOTPRICE suggests that it's also probably not something that we want. It's possibly unencyclopedic, and it's very likely UNDUE emphasis (why that one size, that one date, that one country?), even though it's not actually original research. I don't think that we'll get a sensible RFC response if we go to editors and say "He copied this information straight out of a single reliable source – that's a NOR violation, right?" NOR means "material—such as facts, allegations, and ideas—for which no reliable, published sources exist". If you're copying it straight out of a single reliable source, then it's not NOR. IMO we need to stop talking about NOR (which is either not a problem at all, or is a problem that can be fixed by copyediting) and focus this discussion on DUE.
"Focusing on DUE" IMO means that we ask editors how much emphasis we should put on this subject. "How much emphasis" is partly subjective. Yes, you have to have the sources, but if something is "always" DUE, then you can/should write a weak claim from whatever source you can get.
We need to know whether editors want minimal emphasis (in which case, we remove a number of existing statements, or at least move them out of the lead), or whether they want significant emphasis (in which case, MEDMOS can provide information about how to write non-NOR statements), or something in between.
It would be good to educate the respondents about how complex the subject area is, but if they want significant emphasis on this, then the feasibility of implementing their goals in any given case is not the primary factor in writing our advice. After all, we put a huge amount of weight on the dates and locations of people's births, even though we know that can't source birthdates and locations for every single biography. Putting a huge amount of weight on it just means that if you've got any source at all, even if it's just one unimpressive primary source, then you include whatever you've got, with whatever WP:INTEXT attribution and careful description that allows readers to understand the limits of the source.
When this started, I gave a pair of examples in MEDMOS:
If editors want prices at all costs, we could add one that shows how to respect NOR while still including database-derived prices:
We could also add an example from news media:
I know those statements can be sourced for some medications without transgressing NOR. IMO what we need is to know whether editors actually want us to do that.
With that in mind, I think that the question to ask is:
How much WP:WEIGHT should be put on the prices of pharmaceutical drugs?
This question can (and IMO should) be followed by examples (the despair-inducing table) and explanations (the impossibility of finding the One True™ Price for a drug that sells in 190 countries in six different doses and four common formulations under at least 90 brand names) and alternatives (we could skip dollars-and-cents and instead try to source a comment about affordable/expensive), and comparsions to similar subjects (e.g., how NOTPRICE is applied to other products), but I think that this is the most basic question to ask.
I do not think that we should be asking a yes/no question. I think editors should be encouraged to respond with both an overall view related to inclusion (e.g., always include, usually include, usually omit, only include under extraordinary circumstances, only for generic drugs, only for WHO Essential Medicines, only for drugs under patent protection, only for drugs with unusually high prices – whatever editors actually want) and with an idea of how to include (e.g., brief mention in the lead vs a whole paragraph or section, current prices vs original prices, etc.). WhatamIdoing ( talk) 19:14, 27 December 2019 (UTC)
WhatamIdoing, sorry but I think you are totally wrong, because you keep having a strawman argument about fictitious possible price/cost/affordability statements some fictitious article might contain. We aren't having an RFC because of fictitious possible article text. Of the 500+ drug articles that currently display prices, all of them required original research to make the statement they do. When we chose one unnamed tablet to represent "the cost", that was indeed original research and when we multiplied by a "dose" that was also original research. Have you looked at the medicaid source links? They cite a "prices for week xx in 2018" database of tens of thousands of records. You need to then, by hand, filter the results to the drug the article is about and then you need to look at different formulations and tablet sizes and try to reverse engineer which one was picked to get $50.45 a month or whatever we claimed. And of course the medicaid site doesn't mention a dose at all, so no "intelligent reader" could possibly work it out from that source -- the very definition of original research.
And we also made claims that are not supported by the source at all, like "the wholesale cost in the developing world" citing one buyer price in Costa Rica, or claiming the price ranged from $x to $y when that is also not supported by the source. None of actual articles formulate the price statement like you did. If they did, then you could indeed make the argument that it was undue weight to mention one tablet size from one drug manufacturer in one country. But they don't and the difference is not solvable, as you put it, by "copyediting".
We nearly always give a price per dose, per day or month (which also require picking a dose) or per treatment (which requires an indication and dose). But we always don't mention what that dose is nor do we mention what the indication is. Terbinafine was one example above, Aciclovir another I spotted today, where the article does not state whether the costs are for a cream for cold sores or a tablet for shingles or post-transplant infection of cytomegalovirus. Yes there would be a WP:WEIGHT problem with explicitly giving the price for just one indication, and there would be a WEIGHT problem if we were explicit about the formulation/supplier/etc we used for our maths. But we aren't even specifying the indication, never mind the other things, so our problem isn't WEIGHT but just being dishonest with our readers in claiming there is One Price. The "we could be as specific as The Database is" argument is a false one to make and knock down, because nobody is proposing it. As you say yourself, just giving an example of all the permutations should be despair-inducing. We need to also remember MEDMOS prevents us from stating drug dose information in articles (for good reason) so we can't even explain to our readers why we picked the 250mg tablet.
WhatamIdoing, I have complained about the horrendous problems with the lead text in 500+ drug articles, and I think an RFC that appreciates the problems with that text will result in all those lead prices sourced to databases texts being removed from all 500+ drug articles. Please, the RFC must resolve the conflict over existing article text, not generate imaginary conflicts with imaginary texts. I don't think your WEIGHT question resolves this conflict at all. It isn't even in my mind a sensible question to ask. WP:WEIGHT is determined by reading the body of literature on the article topic, not by Wikipedians expressing a personal opinion. -- Colin° Talk 12:54, 28 December 2019 (UTC)
"The wholesale cost in the developing world as of 2014 is between US$0.36 and US$0.83 per day. As of 2016 the wholesale cost for a typical month of medication in the United States is US$70.50". [18] [19] The DDD on the MSH site says 600mg daily dose, as does my BNF and Drugs.com. So that's two 300mg tablets a day as the developing world prices agree. But the US price of $70.50 is approximately the $70.35 I get by multiplying a 300mg tablet price of $2.34487 by 30. The actual 2016 US monthly price should be $140.69. (Why we have one price per day and another per month is beyond my understanding). Leaving aside that neither MSH nor Medicaid state what the typical therapeutic dose is, for us to do original research on, this is just mathematical incompetence. And it is very typical. I am repeatedly seeing prices citing the BNF that assume a pack of 28 tablets is a "month's cost", totally ignoring that a patient might taken more than one tablet a day. So, the evidence does rather suggest that disallowing original research is a jolly good idea, because we are crap at it. And then we see that the price hasn't been updated since 2016. As your citation shows, the equivalent price in 2019 for 60 tablets would be $46.45, which is about $100 a month less. No small change that, but nobody it seems, is interested in either the price in 2016 being right, or giving the right price for 2019. -- Colin° Talk 21:40, 28 December 2019 (UTC)
In terms of presenting this question, I think it needs a bit of explanation. The straight-up question is "How much weight?", but after that, some explanation is necessary. One way (of many ways) might be to explain the context, and then offer some considerations. It could look something like this:
All Wikipedia articles should present information with WP:Due weight. Generally, this means that the more our reliable sources talk about an aspect, the more attention that aspect should get in the Wikipedia article. However, there is some information that is considered so important that it is included whenever possible. For example, in a biography, we include information about the subject's birthdate whenever possible, but we normally mention the subject's hair color only if reliable sources dwell on the person's appearance.
Drug pricing and affordability is a significant area of discussion in reliable sources, but this discussion is almost always held at a general level, and does not extend down to individual products. The prices of individual pharmaceutical products vary so widely by place, time, dose, and other factors that general claims, such as "the price worldwide" or "the price in developing countries", are almost always incorrect. It is, however, frequently possible to source a statement about what a particular metric yields for the price of a particular size of a particular drug from a particular manufacturer in a single country on a given date.
Editors who work on medicine-related articles have recognized that much of the information about drug prices currently in Wikipedia articles is not an example of our best work. Much of it is outdated or otherwise incorrect. We want to fix it, but we have not been able to agree on the best approach yet. On the one hand, the cost of a drug affects whether people can get it at all, so some editors believe we should always include whatever we can source. Other editors believe that pharmaceutical drugs should be treated like any other manufactured product, and that means no prices unless we have multiple reliable sources discussing the price of that particular product in depth (as we do for some, usually because of very high costs). Editors fall across the whole spectrum from maximizing inclusion and prominence, through the middle grounds, to the opposite side of including as little price information as possible. All of us want to know other editors think, so we're asking you: How much weight should we put on drug prices?
To explain some of the positions, a few editors have offered background information that may be useful to you. We hope that you will join us in a conversation about the best way to handle this subject area.
(Collapsed – It's incredibly important) (Collapsed – You wouldn't believe how complicated and useless this is) (Collapsed – The middle road is WHO/HAI affordability, not prices) (Collapsed – People should care about pricing, not prices) (Collapsed – Where and how we mention prices matter more than whether we do) (Collapsed – What we could actually source is unencyclopedic) (Collapsed – Whatever other ideas/positions/recommendations I've forgotten)
My suggested "collapsed" sections could be written by different people, in the hope that editors would read more than just the headlines. Yes, it's long. That's not necessarily a showstopper. The important question is, if we asked this, do you think that we would get responses that would help us figure out how to clean up these articles? WhatamIdoing ( talk) 06:53, 29 December 2019 (UTC)
SG's attempt at a new start over, abandon hope all ye who enter here
|
---|
IMO, the reason we are unable to formulate an RFC question or questions is that the task we are undertaking is the wrong one: we are attempting to formulate a general RFC to address what is in fact a very specific dilemma relating to very specific databases, when the answers to our sourcing and content dilemma are already addressed by policy. We have no other example anywhere, after weeks of discussion, of any other instances of drug pricing in articles presenting a problem. There has been no problem except the database-style sourcing. We are attempting to generate questions that will get respondents to read and respond to what is (should be) a policy question, but we will get "because I like it" responses that will give us nothing useful as result. That is why we are here. Everyone who has participated in this discussion knows how to add price data according to WP:V, WP:NOR, WP:WEIGHT, WP:NOT and WP:LEAD. Why don't we forget all the general questions we are trying to ask to solve a non-existent general problem, and instead just get straight to the specific problem? Put up one example of database-sourced text (I have repeatedly asked the database advocates to give us the strongest example, and none has been produced) and simply ask if this text is supported according to policy, V, NOR, NOT, WEIGHT, LEAD. Then each respondent will lay out arguments of why it does or does not breach each policy. We are spinning our wheels trying to solve a non-existent problem, as if this has been a generalized problem across all drug articles. We have one problem only; over 500 articles using a database to source text. SandyGeorgia ( Talk) 12:57, 29 December 2019 (UTC) |
Do these examples of pharmaceutical drug prices comply with Wikipedia policies on verifiability, due weight, no original research, and what Wikipedia is not? If so, should this text be inserted into leads of articles?
But the examples need to be varied to included the other drug databases, and the other kinds of problems presented; I pulled these samples from one section above, but the three of them were to demonstrate one issue. The other kinds of examples should be give in place of two of these. SandyGeorgia ( Talk) 14:23, 29 December 2019 (UTC)
I had wondered a similar thing to Sandy's idea: that we take each of the four(?) price-database sources and look at how they are being used and then together reach an agreement about the problems with them, what could be said using them, and what shouldn't be said using them. I actually remain optimistic that much of that analysis/discussion/conclusion could simply be done if we get a good-faith article-experienced bunch of editors to simply work together. All the previous discussion on specific problems felt like it was only me and James and that didn't work for various reasons. I'm not rejected the idea of an RFC, but it would be great to clear away a lot of the crap first, and there really is an awful lot of pretty straightforward crap we could eliminate IMO quite quickly, and focus an RFC on asking the community about price statements that are actually source->text honest and policy-legal. I would be much more confident that such an RFC would be successful in its goals (and to be honest, presenting all the awful prices to the entire community right now would IMO seriously dent WP:MED's credibility).
A mix of family priorities, the latest Star Wars film, and going back to work mean I've not yet finished some of stuff I was working on that demonstrate current text and current problems. I do hope I get a chance to put them up very soon. I think then it will be good if we can all see an honest full selection of drug prices in articles (rather than anyone accuse of picking a hard/easy example) and also quite a number of drugs where the price is genuinely notable and editors could make a really good effort to polish some excellent guideline text on how to present that to readers.
Barkeep49, could we have some kind of moderated workshop to focus on e.g. one source at a time. Advertised to the community. We want participation from wise owls or diligent investigators, rather than just ask for a big mob to vote all at once. I don't think we require medical experts at all, so please nobody rule themselves out on that regard. This is really straightforward source->text analysis. The only kind of ability I can see being useful is an appreciation of statistics to the degree that one can't make general statements from few data-points and to identify the weaknesses in one's data. But that is elementary stuff. We could do this workshop on this page even, and do one source at a time. I would certainly like all the current participants to help, but there is one obvious name who is not currently engaging in discussions, and who's participation is essential. -- Colin° Talk 10:06, 30 December 2019 (UTC)
These examples of pharmaceutical drug prices have been taken from the leads of some articles. Do they comply with Wikipedia policies on verifiability, due weight, no original research, and what Wikipedia is not?
I would like a bit more time to prepare some data pages like I've linked below. There's also a couple of wiki articles on price-related topics that still need a bit of work. I'm keen that we present real data rather than appear to cherry pick. As Sandy notes below, for many of the sources, other editors will not easily grasp how on earth the price was derived from the source. It has taken a while for us to figure this out (while picking jaws up from the floor at the amount of original research and arbitrary choices made). WhatamIdoing you ask "Do they comply...."? Is there anything about this that the group here currently discussing prices disagree on or don't know the answer to? I still think a workshop could be a better approach to tackle and resolve the basic stuff that isn't opinions about what is or isn't encyclopaedic or what does or does not belong in a lead. Those are questions we could ask an RFC, but there are basic mistakes with all the texts & sources that really we don't need to ask the community. Do we? Having an RFC to resolve issues with what one editor alone has written and defends is a bit weird if you ask me. If you ask the above question, it is admitting that WP:MED has no competence to discuss, reach consensus and write honest source-based facts about drug prices. That it hasn't a clue whether the prices in the articles are good or bad and needs some help from Pokemon editors and Historical Fiction writers. And while I think that has been true, that WP:MED has been incompetent here and has failed to address or even examine the problem over many years, it doesn't need to be. -- Colin° Talk 14:24, 31 December 2019 (UTC)
"Having an RFC to resolve issues with what one editor alone has written and defends is a bit weird if you ask me."which takes a jab at an editor (who while unnamed is clear to us who've been following this) and which could have been omitted without diminishing the larger point. The broader community decided that the way forward was through the RfC process, a way of gaining binding consensus. Any consensus reached without an RfC will need to include the consent of editors not currently participating in the discussion but who are invested in the outcome. Unless those editors choose, voluntarily because this is Wikipedia and we all have options afforded to us by being volunteers, to agree to that consensus then it will need to go to RfC. I wouldn't say that the chances are 0 of finding consensus without an RfC but they are slim. And if we can't get to an RfC then the only option is to focus on the behavior issues first (through ArbCom) in hopes that this then creates an atmosphere where consensus can be found on the content question. And even then we still might end up with an RfC. Best, Barkeep49 ( talk) 17:43, 31 December 2019 (UTC)
I've posted a reminder of this discussion at WT:MED#Plans for RfC about drug pricing. -- Tryptofish ( talk) 22:12, 3 January 2020 (UTC)
I have created User:Colin/ExistingPrices that is an automated extract of drug prices from drug articles. I got the list of drug articles by looking for external links to the MSH price guide, the Drugs.com price pages, the Medicaid NADAC pages, or referred to the BNF. It isn't all the drugs, but it 530 is good sized sample. I then extracted lines containing the word "price" or "cost" and did a bit of hand-editing on the result.-- Colin° Talk 17:24, 30 December 2019 (UTC)
<references />
tag at the end of each ===Example===? I'm pretty sure that a simple regex find-and-replace across the page would do it, but I wasn't able to figure out the right combination. The main ref tag was updated to auto-limit itself to only the stuff in between the current one and the previous one, so that will get us the refs in each section (so people are more likely to look at them) without exceeding the template limits.
WhatamIdoing (
talk)
20:01, 30 December 2019 (UTC)
I have also created User:Colin/MSHData which lists every MSH Price Guide reference along with the data year cited and the number of suppliers and buyers. In the WHO/HAI price survey methodology, how representative reference prices are depends on the number of suppliers quoting for each product. Because of this, they focus on a small set (14 or 18 products) that have good supplier data. We can see that 30% of our drug citations have no suppliers at all, yet we claim a price in "the developing world". A further 28% only have one or two suppliers, which makes the claim to be representative of "the developing world" a tenuous one. The majority (58%) of our MSH citations for "the developing world" refer to fewer than three suppliers. While some suppliers are international in scope, many target a single country or even just one part of a country. Nearly all (92%) of our MSH prices are from 2014, five years ago. The remainder are from 2015. The guide used to be updated annually but has not been updated since 2015. -- Colin° Talk 23:14, 30 December 2019 (UTC)
Sandy the prices from the Tarascon Pocket Pharmacopoeia are based on a $, $$, $$$, $$$, $$$$$ pricing symbol much like your holiday guidebook indicates if a restaurant is a cheap-eat or an expensive night out. I complained about it earlier at WT:MED. Here's what they the book says about its symbols: (the underline italics is theirs)
So whenever you see "is inexpensive" or "under $25" that came from a "$". If you see "between $25 and $50" that came from "$$". And so on. If you see "more than $200" that came from $$$$$. So even if the drug costs an eye-popping $9000 a dose, we'll just say "more than $200". And as you point out, the majority of drugs are "under $25" even if actually they are just a few dollars. None of the values 25, 50, 100, 150 and 200 appear in the source-data for the wholesale price of those drugs. Those values are all artefacts of Tarascon's price grouping into $ symbols. Our readers couldn't give a damn about Tarascon's price grouping. Reverse-engineering a $ into "less than $25" is a heinous crime. Saying a thousand-dollar drug is "more than $200" is a heinous crime. -- Colin° Talk 13:58, 31 December 2019 (UTC)
In thinking about how this relates to or affects the general question we've posed, I decided to look at the other med we deal with most commonly in the clinic where I am an interpreter. I was not as familiar with metformin pricing, because we are often giving away free samples. (Actually, many patients have their family send metformin from Mexico.) So, I decided to check that one (diabetes). Wikipedia has:
What remains astounding about this formulaic editing is that, in one demonstrable case where we SHOULD have information in the lead about price ( epipen), there is NONE. We could give that as an example, but we can't edit prices right now. (If we decide we need that as a good counter-example, we can do a mock-up.)
In re-thinking how all of this impacts the formulation of the RFC questions, I am coming back to the lead problem, which must be dealt with.
These examples of pharmaceutical drug prices have been taken from the leads of some articles. Do they comply with Wikipedia policies on verifiability, due weight, no original research, and what Wikipedia is not, and do they reflect the guideline on leads?
What else needs to be done before the RFC is officially launched (other than waiting a couple more days)? WhatamIdoing ( talk) 00:03, 1 January 2020 (UTC)
Here are the ones I put up earlier ... we have evolved :)
SandyGeorgia ( Talk) 00:53, 1 January 2020 (UTC)
I had written a long update to the community spurred in equal measures by how close we are to the tipping point here (either towards success or towards a failure to formulate an RfC) and QuackGuru expressing a desire to appeal an aspect of the prohibition. Quack has now said they're not going to appeal so with half the justification for the update gone I undid it.
Barkeep's Summary at AN
|
---|
A little over three weeks ago I closed a long and unsurprisingly acrimonious ANI thread relating to the behavior of several editors. The ANI discussion also had heavy elements of a content dispute around what should be or not be included in the Medicine Manual of Style page. The major finding was that an RfC was needed. In the time since I (as penance for closing that discussion) have been helping interested editors move towards an RfC. I am updating the community now both because I think we're going to be at a tipping point soon as to whether an RfC will ever get crafted (I'm hopeful but it's hardly a sure thing) and because an editor has expressed desire to appeal to the community for an exemption to one area of that close which I expect will be forthcoming soon. All are of course welcome to read the whole long discussion but here's my summary of major points since then:
While all of the editors participating are incredibly skilled and knowledgeable about the topic – far more so than I – sometimes that understanding of complexity sprawls the discussion in interesting, relevant, and important to the topic ways but not necessarily ways that are helping lead to the RfC. I am hopeful that this update is helpful when considering the appeal that is about to come and in the interests of having some more uninvolved editors who can help move the RfC to launch. Best, Barkeep49 ( talk) 01:08, 1 January 2020 (UTC) |
If this is going to get to RfC the more focus we can have on the specific examples to be used the better. Naming is not unimportant but if we can't finish finding the examples for H/I/J the name won't matter. I suspect that this next week is going to be all the difference as to whether we will get the RfC launched and have it find a consensus (still my hope and well with-in our grasp) or not. Best, 01:32, 1 January 2020 (UTC)
"Sure we could start with "do you think the approximate price of a medication can be estimated for various regions of the world?" And "Should Wikipedia articles contain information about the cost of medications?". Furthermore I think it is relevant that when WP:NOTPRICES was quoted earlier, James claimed to agree with it, despite clearly disagreeing with others as to what it means. I note at Talk:Ivermectin#Price that Seraphimblade wrote
"Pricing, per policy, is not permitted in articles, with rare exceptions when the price is a significant part of something's notability or is very extensively commented on (not just mentioned).". Ronz linked here saying
"All discussion so far supports removal"and James yesterday replied
"Ah lots of discussion supports keeping it.". James's two questions aren't a million miles away from my Question A: can we establish "the price" "for various regions of the world" and "should articles contain them" (though most of us want a "while obeying policy" included in that question). Rather than addressing that bigger question, we seem now to be focusing on chipping away at the prices, either by establishing each source shouldn't be used like it has been (and possibly discover if there is any way it can be used), or get community agreement that the current examples fail policy. Maybe that is the correct approach, but we need to be clear that the chipping-away RFCs are not asking the-big-question. Even if we conclude that our uses of sources A, B, C and D are all awful and should never have been permitted and must be swiftly removed, we still haven't resolved WP:NOTPRICES because some editors read it to mean one thing and others read it to mean another. -- Colin° Talk 11:22, 1 January 2020 (UTC)
I think we have to decide next who is going to be the editor to start chunking text in where. My brain is linear. I am to a point where I am stalled until I see something on a page. And we know we don't want me doing the writing. SandyGeorgia ( Talk) 20:21, 1 January 2020 (UTC)
Barkeep used the formulation of good news/bad news, and I'm going to do my own version of it. First of all, I think it's entirely due to say out loud that we all owe Barkeep a debt of thanks for his very helpful guidance here. Thanks! I also think it's good that we seem to be having a growing consensus that it's a good thing to present editors in the RfC with specific examples to evaluate, as opposed to asking about generalities. And I do think that we are making genuine progress. For me, that's the good news.
Now for the... you know what. As much as we really are moving forward quite well, and despite the fact that we are approaching our self-imposed deadlines, I think that we have yet to resolve some really important issues for the RfC, and we cannot just wish them away. I'd rather get it right, than get it fast. We've been saying that G/H/I/J are getting close to what we want. But I want to be honest about that: I'm not really seeing a consensus that we are there yet.
I said something earlier, and I'm going to repeat it because some editors agreed with it then and I don't think anyone has really objected to it. I've said that the RfC needs to address two issues, both of them in a way that will lead to a clear consensus, one way or the other:
Those are not proposed questions. They are issues that the community needs to answer, in a way that no one will be able to argue against once consensus has been achieved. Even if we get a consensus that, no, we should not be citing drug prices to those sources, we still need a further consensus that, yes, we should present the information this other way. That's important: we need to get consensus for something and not only against something else. And I've also said that I don't think that we can really accomplish that in a single question. I also don't want to leave the RfC format so open that we fail to get focus in the responses.
So: it seems to me that we need to think outside the box, compared with G/H/I/J. I've been thinking about this hard, and it seems to me that we need to present the community with two or more specific choices. For each of those two or more, there should be specific examples of what it would look like on the page, how it would be sourced, and the policy-based rationales for it (or against something else).
If editors here are receptive to that, I can propose what it might look like, but I don't want to do that unless there actually is interest. -- Tryptofish ( talk) 22:16, 1 January 2020 (UTC)
"The wholesale cost in the developing world is about US$27.77 per month as of 2014"
"In 2014, a non-profit organization sold 250 mg tablets of ethosuximide for US$0.1845 each to recognized healthcare organizations in the Democratic Republic of the Congo"
Dear friends, I am unwatching this page because, IMO, our considerable efforts here are being undermined by back-channel conversations, and the topic of drug pricing has not been confined to this remit as required at ANI. A few of us are doing all the work to solve problems that aren't being addressed as the ANI receommended. Sorry, bye. Ping me if there is any urgent need for my useless and verbose opinions. I will keep the RFC formulation page started by WAID watchlisted. Regards, SandyGeorgia ( Talk) 14:18, 2 January 2020 (UTC)
Do you think that this content complies with Wikipedia's standards for verifiability, due weight, no original research, what Wikipedia is not, and how to write a lead sectionis biased. It is the kind of question we ask of content every day. It is a very ordinary question for Wikipedians. It is really very neutral. Are the examples neutral? Well look at User:Colin/MSHData and see if you think they are representative. I think (with the substitution of diazepam for mebendazole) they are. If you think the background and info is biased then please argue specific problems. It is also asking for comments, not a vote, which is something I very very much support. All wiki wisdom suggests a plain vote on polarised options (which is your RFC) is a recipe for polarised and adversarial comments and disunity and in the end "consensus" by counting votes. I have no doubt that some in the pro-prices faction will totally ignore the factual, source and policy issues, and present their case based on Big Pharma suppressing prices and Wikipedia being Not Censored. We will see sweeping statements that of course WHO/MSH and Medicaid and BNF are totally reliable secondary sources. And drug pricing is such an obvious world concern you'd have to be a drug company shill to want to censor prices. And some people will be totally cool with wiki-docs doing original research. This will happen. Let it happen. If Wikipedia decides those things are more important then so be it. -- Colin° Talk 22:16, 4 January 2020 (UTC)
Thanks for the ping, Tryptofish, because I did miss that.
I've been one of the regulars at WP:RFC for about a decade now. I consider accusations of a "biased" question to be par for the course on contentious subjects, and if you want to search the old archives, you will find that my opinion is consistently that claims that "The question is biased!" mean "My side is losing!" I'm not afraid of seeing those accusations, and my plan is to ignore them, because they say more about the fears (and usually inexperience) of the accuser than about the question. (Now, if you personally thought that the question was biased, I'd be very concerned, but that doesn't exactly seem to be your concern.) If you'd like, we could ask the other RFC regulars whether they think the question is biased.
The other theme in your comments is that the results might be hard to interpret. I agree with you: That's a risk. However, I see this phase as information-gathering, and then (as stated in the RFC "question"), I expect the next phase to involve a proposal that people can be for or against.
The way I expect to handle this is to care less during this RFC about summative "votes", and more about the rationales. Let's say, hypothetically, that editors read the three RFC examples and say, "Ugh, that's all wrong". What actually matters to me is the next sentence: Do they think it's bad because per-pill costs were converted to monthly costs? Or is that okay, and the problem is that the sentences are in the lead but not in the body? Or because MEDRS suggests a five-year timeline for sources (all the WT:MED regulars know what I think about that) and 2014 prices are now technically six years old? Or because they think NOTPRICE for a drug requires a source saying that the price of specifically ethosuximide itself is terribly important to the world, and no amount of sources saying that the price of WHO Essential Medicines for epilepsy can possibly suffice to prove that the price of ethosuximide is worth mentioning? Once we've seen which points of policy and common sense people care about, I think we can build on your draft to make an actual proposal for adding a section about prices to MEDMOS. Or, to put it another way, the purpose of this RFC is to make yours produce a clear consensus for fixing this guideline. This is just the necessary baby step to get us on that path. WhatamIdoing ( talk) 01:52, 6 January 2020 (UTC)
;-)
I have more than once attempted to shorten something and ended up with a significant increase in the length. I can simplify, but shortening is not my strength.
Tryptofish I appreciate that asking only about MSH-sourced text will not reliably put out the fires at NADAC or BNF or Tarascon or Drugs.com sourced texts. But if anything can be seen from the volumes I've written about the flaws in our texts, is that it is amazing that one can make so many mistakes and commit so many policy crimes in just a few words. I think if we try to explain why the "drug prices everywhere" approach is not just not-encyclopaedic but also totally impractical, for all examples, we will get totally bogged down. Fatigue will then likely to see over-simplistic replies like "Support: we using secondary and government sources for prices such as Medicaid, MSH, and Tarascon". We need to allow participants a chance to focus on a smaller level of practical issues than "all drug price sources".
I think Wiki largely discourages "prices everywhere" on fundamentally practical grounds, more than on "encyclopaedic" grounds. We aren't a standard paper encyclopaedia and already contain much trivia or dubious lists of facts. Practically, there are plenty sites (GoodRX, Drugs.com, BNF in the UK) where patients and readers can reliably find out about drug prices in their country, and get figures that are accurate TODAY for the indication or prescription they are concerned about, rather than citing a book from 2015 for a totally unknown indication and dose. All the evidence suggest Wikipedia is crap at this. Same goes for the prices of other things from mobile phones to insurance to properties in your area to train tickets.
I think it is a huge mistake to frame drug prices round concepts like primary and secondary sources, which Wikipedia has historically had mixed interpretations of meaning. James says above that the BNF is a secondary source, and it is true that it gets its information from the NHS, who in turn give a mix of regulated price and/or indicative price for the drugs, the latter of which comes from drug companies and pharmacy contracts. A better distinction is that all those sources (with the exception of Tarascon's $$$ symbolic prices, which have their own huge problems) are simply databases of prices of products with barcodes, and all those sources give their own unique kind of price (all different variants of wholesale or retail, some actual, some list, etc). Those sources are raw data, at a level of multiplicity and complexity that none of us are proposing is reproduced on Wikipedia.
Wrt notability of prices, saying "secondary sources have written about issues that are specifically about the pricing of those medications". Repeatedly it has been claimed that the fact that e.g. BNF include prices for all their drugs (similar for Drugs.com and Tarscon's book) means they have been written about. Clearly the authors of those sites/books, when writing about each drug, considered it relevant to give price data. And Google will find someone somewhere mentioning that X is a low cost drug compared to the new drug Y which is expensive. But doh!, all new on-patent drugs are expensive and most existing generic drugs are cheap, so that isn't exactly news to anyone that someone might make that remark in print or online. The advocacy argument for including prices is so strongly held by some, that we need a much higher barrier-to-entry than simply being written about in secondary sources.
I'm very nervous about get-out-of-jail clauses like "or used only with care". We aren't here to redefine fundamental policy. Anyone can argue they are careful. We see in the lead RFC that e.g citation excess is justified on the grounds that there is no policy against citations: any guideline-caution or recommendation to seek per-article consensus about citations is simply cast aside. An "or used only with care" clause simply says one can ignore the preceding text.
So I don't think, sorry, your RFC is appropriate today. Let's start with the RFC on MSH-sourced price statements today and see where that takes us. We can learn lessons from it. It may be that Wiki so clearly rejects raw-database-sourced drug prices that existing policy on WP:NOPRICES becomes the clear consensus, and we all do already know how to write about prices when newspapers, etc have made comments about them and give us a price-to-treat or a price-per-year without us having to get our calculators out. -- Colin° Talk 08:54, 3 January 2020 (UTC)
If there are any "back door" discussions, I'd like to know where they are. ... --Tryptofish (talk) 23:36, 2 January 2020 (UTC)Putting this here per your request, SandyGeorgia ( Talk) 23:38, 3 January 2020 (UTC)
User:Tryptofish thank you for the notification at WP:MED. I was not following the debate but would like to participate as a newbie. As I understand it, there area two parts of the debate. The first is regarding point #5 of the WP:NOTDIRECTORY which states,
"Sales catalogues. An article should not include product pricing or availability information unless there is an independent source and a justified reason for the mention. Encyclopedic significance may be indicated if mainstream media sources (not just product reviews) provide commentary on these details instead of just passing mention. Prices and product availability can vary widely from place to place and over time. Wikipedia is not a price comparison service to compare the prices of competing products, or the prices and availability of a single product from different vendors or retailers."
From what I can read, part of the RfC will debate whether or not medication prices falls into the exceptions listed above (e.g. is there a justified reason). The remaining question will debate which medications and which sources are acceptable? Can you confirm if I have this correct and if so, is the draft on your talk page now? Also which opinions you're seeking at this time. Thank you for shepherding this topic. Ian Furst ( talk) 15:06, 4 January 2020 (UTC)
User:Barkeep49, I think Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices is ready. As I understand it, ANI thought it would be best if an uninvolved admin endorsed its neutrality. Can you post a request at some suitable forum to find volunteer for that step?
Here are my thoughts about how to manage the RFC once it's underway, and I'd like to hear all of yours, too.
First, I'd like to encourage all the "principals" in the original dispute to stand down for the first day or two. Let's imagine that an admin certified it as neutral very soon. In that case, I could probably take the draft tag off and list it as an RFC as early as Tuesday. If that's the schedule we end up on, and if you think that other people might think you've got a dog in this fight, then please stand down until at least Wednesday. There's nothing like long or angry posts, especially from our recognizable community leaders, to scare off some contributors. I may not post my own views at all, and I certainly don't plan to do so during the early days. We can let this run for weeks, or even months if we're still getting good responses. There's no magic timer for RFCs. I promise that you will have a chance to have your say, even if your post is #10 instead of #1.
Second, I want us to be encouraging the uninvolved editors to engage in this RFC enough to tell us what they think. Some people will just want to dump a drive-by vote on the page, but if they're willing to explain their thought process, then I want to find out more. I am discouraging straw-poll or "survey" approaches, and I hope that you can all support that in practical ways, like breaking long discussions into sections with useful names like === Thoughts on X ===. Getting detailed explanations from our volunteer editors is a gift that we should treasure. I expect to be asking some editors questions to encourage discussion. If you think that you can ask a question that will draw out more details from an editor or that will encourage that editor to connect with another editor, then please consider doing that. Something like "Do you feel like your idea relates to what User:Example was saying last week?" or "Do you think that might work better with <this slight change>?" could be good. The goal is to get the other guy talking. A good, responsive question, phrased with respect, can be an excellent tool for producing further explanation. An amazing success looks like a couple of editors putting their views together to come up with something that's better than what any of them started with.
Third, if you see an opportunity to meatball:DefendEachOther, especially if it's someone from the other "side", please do so as quickly and as gently as you can. Nobody involved in constructing this RFC wants Wikipedia to get worse. We all have the same ultimate goal. If you'd like, I can ask the WT:RFC regulars to help out with this.
Fourth, I personally don't feel like I'll need an official "closing statement" to know what I've learned from this RFC. However, if you do, then please be bold and speak up now, especially if you'd like to have a "team of three" approach. Recruiting three people after an RFC has ended can be difficult and result in needless delays.
If others have advice they'd like to add, or would like to suggest a different approach, please post here. I'll make time to check this page between meetings (probably in ~12 hours or so). WhatamIdoing ( talk) 06:41, 6 January 2020 (UTC)
"Buyers: These prices should not be used as international reference prices". WHO/HAI says:
How representative reference prices are generally depends on the number of suppliers quoting for each product.. We do not have a source saying "IDA is representative of wholesale prices in the developing world".
"in many low- and middle-income countries medicine prices are high, especially in the private sector (e.g. over 80 times an international reference price); availability can be low, particularly in the public sector (including no stocks of essential medicines); treatments are often unaffordable (e.g. requiring over 15 days’ wages to purchase 30 days’ treatment); government procurement can be inefficient (e.g. buying expensive originator brands as well as cheaper generics); mark-ups in the distribution chain can be excessive; and numerous taxes and duties are being applied to medicines"they give an example:
"The price of originator brand atenolol 50 mg tablets is over 20 times the international reference price in all the countries except India (where it is still high at 5 times the reference price) and Kazakhstan. Even the lowest-priced generic is very expensive in all countries". So basically, the wholesale price of generic medicines to the state healthcare is irrelevant if the state healthcare has no stock of that medicine and the patient has to buy a premium-brand version on the private market. And that is the norm.
Quick replies:
WhatamIdoing ( talk) 16:40, 6 January 2020 (UTC)
Just a note to all involved that I decided to take a break from this yesterday and am catching up on it today. I want to acknowledge that I've seen WAID's comment above but want to have caught up fully before I launch. As I am fairly busy at work this week, I may not be able to fully catch-up here until this evening. Just wanted to set appropriate expectations.
Barkeep49 (
talk) 17:19, 6 January 2020 (UTC
I wish to add a link to Wikipedia:Prices#Discussions_about_best_practices to the RFC on pharmaceutical drug prices. This link presents a list of all previous discussions of drug prices, which I feel match the subject of this RfC. The point of sharing links to this previous discussions would be to show the history of Wikipedia community discussion of drug prices.
SandyGeorgia objects, saying with a revert that "this is not an RFC on pricing, this is an RFC on source --> text integrity". What reason is there to avoid presenting this archival collection of previous discussions? Blue Rasberry (talk) 15:26, 7 January 2020 (UTC)
IMO, the editors representing the polar opposite lead "sides" of this discussion should refrain from directly editing the RFC. I have said this before, but now James is adding contentious statements or replacing/removing factual information with over-simplified statements. He has disputed that prices are "incorrect" and has replace the
with
The problem with, what WHO admit is simply a "basic definition" is that it is over-simplistic and misses out the key factor of DDD that our article gets right: it is a statistical measure of drug consumption, a rough estimate for population-level studies. This statement implies this might be an average Prescribed daily dose and it really really isn't. Let's look at what WHO say:
So in practice, the figure is much more complex and involves a human making a non-clincical judgement. Hence the original text said "a complex statistical concept".
The
WHO also say: "The main purpose of the ATC/DDD system is as a tool for presenting drug utilization statistics with the aim of improving drug use. This is the purpose for which the system was developed and it is with this purpose in mind that all decisions about ATC/DDD classification are made. Consequently, using the system for other purposes can be inappropriate."
. The DDD is also not updated to reflect changes in prescribing practice or recommendations. A better definition would be (from WHO)
So I think we should stick to linking to the wiki article, and reverting back to the original text which made a neutral statement about this technical metric rather than offering an over-simplistic statement that will mislead. Our readers who will wrongly think that "average maintenance dose per day" reflects some clinical-practice average prescribing statistics rather than some pen-pushing statistician looking at drug dose remarks in the BNF or wherever, and choosing either the initial, the max dose or the mid-point between these depending on the words they read. It is simply "a complex statistical concept", a "technical metric", with no clinical application whatsoever, and we should not pretend it is in any way related to actual clinical practice.
James has repeatedly stated a view of DDD and how it can be used that is at odds with what WHO use it for and that WHO firmly discourage. For that reason, he should not be the editor to make claims about it in the RFC. He is welcome to state his personal beliefs in the response to the RFC.
Wrt the "Many of these are from 2014 or 2015 or are otherwise incorrect" tagged as "dubious", we have megabytes of prose outlining how incorrect these prices have been. Not just in ways that editors may disagree about but simple maths and statistics errors. I don't think that is dubious at all. Below we had a discussion about Carvedilol where for over a year we have incorrectly stated the "wholesale cost per dose is less than 0.05 USD". If fact there carvedilol ER tablets costing $6.44, $6.61, $7.08 and $6.57 for each 10, 20, 40 and 80mg doses. -- Colin° Talk 13:21, 9 January 2020 (UTC)
"The MSH International Medical Products Price Guide is a generally unreliable source of information about drug prices, because WHO only regard an "international reference price" as being "representative" if it has many supplier records, and many drug records in this database lack any supplier records and many others only have one or two". But we're not doing that approach in the RFC and I'm not about to start edit warring with others to push for that in the Background section. James has contentious views on what DDD is and how it can be used, and this is at the heart of the dispute. Therefore he should not be permitted to write about DDD in the RFC in a way that tends to support his views. If you value his and my opinion, you can wait for that in the response section of the RFC. Does that not seem reasonable? -- Colin° Talk 15:30, 9 January 2020 (UTC)
James (can someone else please ping him) you are upset about the "Many of these are from 2014 or 2015 or are otherwise incorrect". I am also upset that DDD is not currently being described as "a technical drug use metric" which is also a definition from WHO, and very much gets to the heart of its primary purpose. Instead we use a definition which, without supporting explanation, looks like the definition of average Prescribed daily dose or Recorded Daily Dose, and it very much is not. So I propose a compromise. If you will accept changing:
the defined daily dose ( "the assumed average maintenance dose per day for a drug used for its main indication in adults"; not necessarily the dose any person takes, especially when the same drug is prescribed for multiple medical conditions).
to
the defined daily dose (a technical drug use metric)
Then I'll not complain if "Many of these are from 2014 or 2015 or are otherwise incorrect" gets dropped. Our readers can work that much out for themselves. Obviously WhatamIdoing and others may have their own opinion on acceptable compromises. -- Colin° Talk 10:33, 11 January 2020 (UTC)
The RFC assertion that many prices are "incorrect" has been challenged by Doc James and AlmostFrancis. Of course, IMO the prices are incorrect on so many levels it is hard to know where to begin. But apparently even high-school rules of statistics are open to question on WP:MED so I'll stick to listing below mistakes of basic maths or where unsourced statements about "dose" are clearly wrong per reliable sources. And I'll stick to using the same methodology wrt cost per DDD that James has used (which itself is contestable). I've only picked 6 examples from the 31 drugs with the letter A that cite MSH (322 drugs total). So that's one in five with clear mistakes:
-- Colin° Talk 18:33, 11 January 2020 (UTC)
The original RFC assertion that many prices are "out of date" and even now mentioning they are from 2014/2015 has been challenged as biased by Doc James and AlmostFrancis. I have compared the unit prices in the MSH price guide cited for 2014 and compared with 2009, five years previously. This can give us an indication perhaps of how much prices change in five years since 2014 data. As above, I'm just looking at 31 drugs starting with the letter 'A'.
So that's 50% (16 out of 31) of drugs where one of the price range figures has doubled or more, or halved or more at a five year interval. And three drugs with extraordinary shifts. Us Wikipedians should be evidence and sourced-based in our assertions and claims. Disagreeing with a statement because it doesn't suit one's case is not enough. -- Colin° Talk 18:56, 11 January 2020 (UTC)
I have also compared the prices for the following year (or in a few cases, the previous year as there was no following year data).
So that's 25% (8 out of 31) of drugs where one of the price range figures has doubled or more, or halved or more, from one year to the next. And two drugs with extraordinary shifts.
In summary, annually we see a quarter of drug prices change by a factor of 2, and over five years we see half of drug prices change by a factor of 2. Some prices will change by huge amounts. We have seen claims that drug prices don't vary that much, and this research shows conclusively that this is untrue. Further, nearly all prices change annually by an amount that makes it hard to justify quoting a price to 4 significant figures. Using the median supplier price where there are three or more suppliers would result in less variation, though that would permit only a small minority of articles to quote "developing world" prices. -- Colin° Talk 13:29, 12 January 2020 (UTC)
Why not include a pro / con statement at the end of the RfC? In the United States these statements come with government voter guides before the election. In my view, the text of the RfC is not easily understandable by anyone who is not already invested in the issue. I question whether this text could inform anyone of the complexity of the issue if they are not already informed. Elsewhere there was wish that the regular players not immediately jump into the discussion in hopes that early voters would post original comments rather than ride in support of the status quo positions.
The people who are already in this discussion have chosen the best identified arguments for presenting their perspective. In conventional debate the various sides put forth their best interpretations. Let's include positions in the RfC because that seems like the conventional way to do this. What reasons are there to not do this? Blue Rasberry (talk) 16:01, 9 January 2020 (UTC)
Ian_Furst and Bluerasberry, please carefully read WAID's carefully crafted RFC and also her recommended approach at the top of this ready-stead-go section. She's explicitly asking for new voices, new thoughts and for them to be based on looking at a small set of representative examples of article text using one database source. What I believe she isn't looking for is a dump of James and Colin's opinions and then for the community to discuss our opinions, biased on their impression of whoever may or not be a World Expert on Healthcare Information Sharing :-)!
I don't believe the RFC is hard to understand:
Main question:
Do you think that the content in the examples above complies with Wikipedia's standards for verifiability, due weight, no original research, what Wikipedia is not, and how to write a lead section?
Our policies require simply that an intelligent non-expert reader be able to look at our article text and look at the source given (and only that source) and determine that the article text is supported by the source. It is a little more work for a reader to determine WP:WEIGHT for that requires one to invest in reading the body of literature on a topic, but it isn't rocket science to consider if choosing one particular (unspecified) strength and formulation of the drug and one (unspecified) indication in order to quote a treatment price is reasonable or not. Our lead guideline also gives advice on contents of the lead vs body. We really really aren't asking folk to have to read all this lengthy debate. Only to examine as you would any wikipedia article and source. Rather than opinions, perhaps I could offer some questions?
Perhaps you can think of your own. They aren't hard questions, or ones that require hours of study or a degree in medicine. -- Colin° Talk 20:15, 9 January 2020 (UTC)
(BTW I'm not wanting answers to those questions here. That's the point of the RFC.) -- Colin° Talk 20:33, 9 January 2020 (UTC)
Do you think that the content in the examples above complies with Wikipedia's standards for verifiability, due weight, no original research, what Wikipedia is not, and how to write a lead section?
I feel the drafts are too long and complicated to get meaningful input from a broad range of editors via an RfC, so I've created a very simple version for consideration: User:Levivich/Drug prices RfC draft 3. – Leviv ich 20:32, 10 January 2020 (UTC)
I don't feel like this RfC has to singlehandedly solve the problem with drug prices
. It doesn't have to be
The RfC to End All RfCs, it doesn't have to be
One RfC To Rule Them All. It does have to be digestible by dozens of editors who know nothing about this. Instead of the heart of the dispute
, pick one important part of the dispute (lungs, brain, really any major organ will do), and resolve that. Then, pick another. Like all good RfCs, it should be phrased in such a way that when it is over, someone can make an edit (or is prohibited from making an edit), and everyone will know that this edit does (or does not) have consensus based on such-and-such RfC. –
Leviv
ich
22:48, 10 January 2020 (UTC)
(sigh). Levivich, I don't think it is helpful at this stage of a multi-month discussion on formulating the RFC, to turn back the clock and basically start again. The RFC proposed by WAID is the result of this process. We all had a chance a month a go to propose "our RFC question" and the current text came about from thinking about and discussing that for a long time. Let's not begin 2020 with "I want my question" "No, I want my question". I'm sure everyone has an opinion how they'd do it if they had their way, but please respect the process we have all been through and the result of that process is Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices. -- Colin° Talk 10:24, 11 January 2020 (UTC)
Levivich, you've put a lot of work into these drafts, but I come back to the simplest possible format put forward so far is the one put together by WAID, and proposed at AN. It is true that editors outside our realm may not engage, but I do not think your drafts spell out to them what they are engaging, or draw them in and promote interest, as WAID's version does. If we complicate too far, we get no feedback. If we simplify too much, we get useless feedback (your comparison to MEDLEAD RFC asking a simple yes/no question). I hope we can get WAID's version moving, so that we can put these source-->text issues behind us and move on to asking the real question of where the community stands overall on NOTPRICE. IF no one engages WAID's RFC, we are at least not further behind, and it is preferable to have people not engage than to have people not understand what they are engaging. I think WAID's version gets to the heart of that matter: how can these sources be used. SandyGeorgia ( Talk) 20:23, 12 January 2020 (UTC)
A problem with this so-called simple RFC is the Stranger Things factor. Anyone coming along to this will see a question like "Should articles state drug prices sourced to drug price databases such as the International Medical Products Price Guide" and immediately remember that they still have two episodes of series three of Stranger Things to watch. They have no idea what the International Medical Products Price Guide is, but the name sounds impressive so it's probably ok. Are any articles doing this at the moment, or is this a proposal to do something new? No idea. Not sure I care then. Should we source drug prices to drug price databases? Well duh, they are drug price databases. What's the issue with that? The question says "such as" but doesn't tell me what the others might be, or if they are different in some way I should care. The second question 3B immediately has me playing spot-the-difference then I notice "wholesale prices in the developing world". Ok, it's a database of wholesale prices in the developing world. What's the issue with that? And 3C has the same question but replaces "wholesale prices" with "any pricing information". But the examples are the same, all stating wholesale prices, so I'm not seeing what different "pricing information" there might be to consider, other than "wholesale prices". What's the issue with that? I'm being asked my opinion on something I don't currently have an opinion on, and don't really understand why I should care at this point. But look, someone whose name I recognise is saying that the Wikipedians who want to remove prices are helping Big Pharma suppress price information, which kills people. And Wikipedia is Not Censored. Ok I care about that, let's Stick it to The Man, yeah. "Support per ____" and fade in the Netflix logo and some 80s-style theme tune.
Compare this with Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices. Immediately we know there are articles with these prices. And I'm being asked "Do you think that this content complies with Wikipedia's standards for verifiability, due weight, no original research, what Wikipedia is not, and how to write a lead section?" Right, I'm a Wikipedian. I can do this. This is something I actually know something about and care about. I'm gently introduced to the source being considered and offered a few examples so my brain doesn't overload. I can see that using this source isn't simple [we are here for months discussing this since it isn't simple]. I don't even, at this point, need to have an opinion about drug prices, just care about Wikipedia's core policies and values. Pass me a slice of pizza, this sounds worth discussing. Netflix can wait. -- Colin° Talk 20:53, 13 January 2020 (UTC)
(outdent) If it would help, we could move the two pages AlmostFrancis complains about into either project (sub page of WP MED) or Wikipedia (sup-page of this) namespaces. The information contained is not personal in any way, and claims they are biased are founded solely on personal attacks of the creator of the pages. -- Colin° Talk 11:23, 14 January 2020 (UTC)
I note that Levivich created their replacement draft, because they felt WAID's was "too long and complicated" and that an RFC "does have to be digestible by dozens of editors who know nothing about this". Levivich wants to ask "should we use the source at all" rather than "how to use the source". But this does not get to the heart of the dispute. Nobody is saying the price guide must never be used at all (though it being a dead parrot suggests its further use is limited) and nobody is saying similar drug databases should never be used at all. I feel most of Levivich's many questions are in fact ones nobody is asking, and don't need asked. IMO the heart of the price problem isn't one of "should" but "can" and "when" and "how". Should is the weakest argument of all, and can be based on opinion and advocacy as we have seen. As I have said before, Wikipedia is full of trivia and editors will tend towards including data that somebody feels is important. If drug prices were stable, were international and were simple enough that there is one obvious treatment cost per drug, we may not be having this argument at all. The problem is practical. It is a can, when and how problem. All the questions in Levivich's ask should we do something, without getting the participants to wonder how. If we want "editors who know nothing about this" to engage we need to give them the background that WAID's does.
It is ironic that Levivich has previously stated they are aligned with WP:NOTPRICES in thinking drug prices should only be included when exceptional and yet their RFC is supported by two editors who appear to want drug prices universally. In contrast WhatamIdoing has for a long time wanted financial information in medical articles but clearly wants it done right, rather than any old way that might break policy. Rather than asking a polarising "should we do this ever" question, they way to find consensus is to agree to do it right, and on a case by case basis, doing it right might mean often avoiding one source that has limitations, or wording our articles to make those limitations clear. I urge Levivich to reconsider: your "simple RFC" will be hijacked with opinionated global price concern issues if it were to be launched, and you will not get anyone even considering the result you want. -- Colin° Talk 11:23, 14 January 2020 (UTC)
Levivich's many questions are in fact ones nobody is askingis a feature not a bug :-) I'm not sure what you mean by
I urge Levivich to reconsider. There isn't consensus for any of my drafts, so I don't see what there is to reconsider. I still think WAID's RfC fundamentally asks the wrong questions (and that was the reason for my proposing alternatives in the first place), and I still think that an RfC that solicits discussion, as opposed to an RfC that solicits a decision, will not bring us meaningfully closer to a decision. Most everyone here disagrees with me on this. The only way to find out is to run an RfC. And with all due respect, I just had this exact same conversation with WAID a few days ago, above in this very same thread, so I feel we're kind of running in circles here. My humble suggestion at this point is to move anything objectionable in WAID's RfC (the background, the links, whatever it may be) out of the RfC question and instead present that information in the first few !votes (which the editors here will undoubtedly be casting), if only so as to move the ball forward and get to a point where we launch an RfC that has consensus among editors here. I don't know whose "law" it is, but we are proving the law that as discussion continues, the probability that we lose editors to retirement or sanctions approaches 1. It's clear the cavalry is not coming, so let's address whatever issues are left with WAID's RfC and get on with it. – Leviv ich 19:23, 14 January 2020 (UTC)
Draft RFC at: Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices
I think we have reached a stage where we should now look to complete polishing Wikipedia:Manual of Style/Medicine-related articles/RFC on pharmaceutical drug prices ready for launch. The concerns raised above are:
User:Colin, I think that the idea of "Is this encyclopedic?" would probably end up in the WP:NOT-related section, but it could also fit into the subject-matter section or possibly the style section. I'm not convinced that further divisions are really warranted unless we get a huge number of comments. WhatamIdoing ( talk) 20:28, 17 January 2020 (UTC)
An uninvolved admin has commented on the draft RfC over at AN: Special:Diff/936639234 – Leviv ich 03:11, 20 January 2020 (UTC)
"You can see several hundred examples of how drug prices are being presented"or even just " What we've got on wiki? If one of both of these feels neutral enough to you, is there a version of In the real world that could also be made neutral? Or are you suggesting the only way for the RfC to be neutral is without these sections no matter their form? Best, Barkeep49 ( talk) 00:15, 21 January 2020 (UTC)
Is this the right place to note some concerns I have about some of the phrasing in the proposed RfC question? If not, please feel free to move them to the appropriate place.
"and none are comprehensive". I find that an unsupported judgement. Databases generally have a fixed range of content, so whether or not they are comprehensive isn't an issue. They either contain reliable information or they don't. I worry that the phrase places the databases in an unduly negative light.
"Editors have raised concerns about prices being outdated or having other problems", while true, it doesn't give the reader any context. From what I can see this refers to about a dozen editors over the space of several years. What about all the editors who are not concerned? don't they get a mention? What of those who think that having the latest prices available is as good as we can get on Wikipedia. Atre their views not pertinent to the background that frames the RfC?
"standards": does Wikipedia have standards written down? I know we have policies, guidance, best practice, and advice, but I don't think we have written standards. I worry that the use of the word will create a non-neutral expectation that our guidance has to meet unrealistic standards. The whole point of supplementary guidance is that it caters for the exceptions to more general policy and guidelines. I worry that it leads the respondent to make a judgement appropriate only for general guidelines, not for subject-specific ones.
"how to write a lead section": are we going to measure potential guidelines that will apply to the whole of a medical article against a guideline that is only concerned with the lead section? Or is this RfC intended only to apply to the issue of drug prices in the lead? In that case, surely the background should clarify that arguments against drug prices in the lead should not be predicated on information in the lead not appearing in the body, as that is already covered elsewhere? -- RexxS ( talk) 17:44, 21 January 2020 (UTC)
lists drug price information for WHO Essential Medicines" according to the lead of our article. If you look at the list of essential medicines, do you see any missing from the price guide? If not, then the database covers its intended content comprehensively. I haven't found any missing, but of course, you might. Nevertheless, the concern I have is that comprehensiveness is irrelevant for a source. Either we accept the source as reliable or we don't. Being comprehensive has no bearing on that. Of course we could only use the database for drugs that it is deemed to have reliable information on, but again, comprehensiveness has no bearing on that. Apologies to all who've heard the argument before, but I was asked. Do you really want me to explain my other concerns in detail as well? -- RexxS ( talk) 21:18, 22 January 2020 (UTC)
"Editors have raised concerns about prices being outdated or having other problems". I contend that without context, it places the use of prices in an unwarranted negative light. Yes, some editors have raised concerns about some prices, but if you examine User:Colin/PriceEdits #Drugs, you'll see that the majority of prices have not had any concerns raised. Of those where concerns were raised, only one was a concern about outdated prices, although one other price was updated to a more up-to-date one. I'm concerned that the statement in the proposed RfC doesn't neutrally reflect the evidence.
User:Barkeep49, I'm still hoping to figure out what the problem with the "real world" section is. Surely it's neutral to point out, as background information, that regular strength and extra strength don't cost the same, or that many prices are confidential. The editors expressing concern about it being non-neutral have been fairly vague about what, exactly, is non-neutral about it.
RexxS, being "outdated", if you apply MEDDATE's five-year standard, affects about a third of what's in articles and about 90% of what's cited to the MSH database. The category of "other problems" includes:
So you're right when you say that "some editors have raised concerns about some prices" before this discussion started, but it would equally be true to say that some editors have raised concerns about nearly every drug price in the entire wiki during this discussion, and that some of these points are being raised by uninvolved, non-medical editors ( example). Off hand, there are probably no small-molecule WHO essential drugs whose current price content (or at least some of it) doesn't fall afoul of at least one of those complaints in some fashion.
For the record, I don't agree with all of the concerns that have been raised. Even among concerns that feel valid to me, I believe that many can be fixed without removing all of the content about prices.
Also for the record, the list of concerns and the (very approximate) percentages given above are from my memory of previous discussions. I've probably forgotten some of the concerns. Some of the numbers I give might be wrong, and others apply only to relevant subsets (e.g., MSH-citing articles, or the most recent year's date in the MSH database). But I do not think it would be fair to describe the concerns as only applying to "some" of the prices. I wonder if, having seen these estimates of how many articles could be affected, you would still describe it that way. WhatamIdoing ( talk) 01:35, 23 January 2020 (UTC)
It looks like this is truly set to go. I know SandyGeorgia has some concerns about the discussion sections. Sandy is this worth holding the launch of the RfC? I am tenatively planning on launching this in 4 hours (23:00 UTC). Best, Barkeep49 ( talk) 19:02, 23 January 2020 (UTC)